# silvesterm05/alpr-using-yolo architecture

> Architecture diagram of the GitHub repository silvesterm05/alpr-using-yolo, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-05). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/silvesterm05/alpr-using-yolo
- Repository: https://github.com/silvesterm05/alpr-using-yolo

## Overview

This repository combines two vehicle-image download scripts with an ALPR character-detection and crop-export workflow. The sampled `seperation.py` localizes plates, detects and orders characters, then exports labeled crops and coordinate CSVs; `final.py` separately detects characters in input images and exports filename-labeled crops. The README describes an EfficientNet classification stage, but no classifier implementation is visible in the supplied files. No source evidence establishes that downloaded images are automatically fed into either processing script.

Main entry points: [`download.py`](https://github.com/silvesterm05/alpr-using-yolo/blob/main/download.py) (Filtered downloader), [`marinaimg.py`](https://github.com/silvesterm05/alpr-using-yolo/blob/main/marinaimg.py) (Mall downloader), [`seperation.py`](https://github.com/silvesterm05/alpr-using-yolo/blob/main/seperation.py) (Plate localization), [`final.py`](https://github.com/silvesterm05/alpr-using-yolo/blob/main/final.py) (Character detection)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_ingestion["Image ingestion"]
  node_download_job["Filtered downloader<br/>[download.py]"]
  node_marina_job["Mall downloader<br/>[marinaimg.py]"]
  node_download_store[("Downloaded images")]
end

subgraph group_alpr["ALPR processing"]
  node_input_images[("Input images")]
  node_plate_localizer["Plate localization<br/>[seperation.py]"]
  node_plate_cropper["Plate crop and upscale<br/>[seperation.py]"]
  node_character_detector["Character detection<br/>[seperation.py]"]
  node_final_detector["Character detection<br/>[final.py]"]
  node_efficientnet["EfficientNet classifier"]
end

subgraph group_outputs["Exported data"]
  node_character_exporter["Labeled character crops<br/>[seperation.py]"]
  node_final_exporter["Labeled crop export<br/>[final.py]"]
  node_coordinate_csv[("Coordinate CSVs")]
  node_plate_string["Plate string"]
end

node_operator(("Operator"))
node_vehicle_api["Vehicle API"]

node_operator -->|"runs"| node_download_job
node_download_job -->|"requests records and images"| node_vehicle_api
node_download_job -->|"writes images"| node_download_store
node_operator -->|"runs"| node_marina_job
node_marina_job -->|"requests records and images"| node_vehicle_api
node_marina_job -->|"writes images"| node_download_store
node_operator -->|"provides images"| node_input_images
node_input_images -->|"reads images"| node_plate_localizer
node_plate_localizer -->|"crops detected plate"| node_plate_cropper
node_plate_cropper -->|"detects characters"| node_character_detector
node_character_detector -->|"orders and exports crops"| node_character_exporter
node_character_exporter -->|"writes coordinates"| node_coordinate_csv
node_operator -->|"provides images"| node_input_images
node_input_images -->|"reads and upscales"| node_final_detector
node_final_detector -->|"exports filename-labeled crops"| node_final_exporter
node_final_exporter -->|"writes coordinates"| node_coordinate_csv
node_efficientnet -.->|"classifies characters"| node_plate_string

click node_download_job "https://github.com/silvesterm05/alpr-using-yolo/blob/main/download.py"
click node_marina_job "https://github.com/silvesterm05/alpr-using-yolo/blob/main/marinaimg.py"
click node_plate_localizer "https://github.com/silvesterm05/alpr-using-yolo/blob/main/seperation.py"
click node_plate_cropper "https://github.com/silvesterm05/alpr-using-yolo/blob/main/seperation.py"
click node_character_detector "https://github.com/silvesterm05/alpr-using-yolo/blob/main/seperation.py"
click node_character_exporter "https://github.com/silvesterm05/alpr-using-yolo/blob/main/seperation.py"
click node_final_detector "https://github.com/silvesterm05/alpr-using-yolo/blob/main/final.py"
click node_final_exporter "https://github.com/silvesterm05/alpr-using-yolo/blob/main/final.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_download_job,node_marina_job,node_download_store toneBlue
class node_input_images,node_plate_localizer,node_plate_cropper,node_character_detector,node_final_detector,node_efficientnet toneAmber
class node_character_exporter,node_final_exporter,node_coordinate_csv,node_plate_string,node_vehicle_api toneMint
class node_operator toneIndigo
```

## Components

### Image ingestion

- [Filtered downloader](https://github.com/silvesterm05/alpr-using-yolo/blob/main/download.py): `download.py`
- [Mall downloader](https://github.com/silvesterm05/alpr-using-yolo/blob/main/marinaimg.py): `marinaimg.py`
- **Downloaded images**

### ALPR processing

- **Input images**
- [Plate localization](https://github.com/silvesterm05/alpr-using-yolo/blob/main/seperation.py): `seperation.py`
- [Plate crop and upscale](https://github.com/silvesterm05/alpr-using-yolo/blob/main/seperation.py): `seperation.py`
- [Character detection](https://github.com/silvesterm05/alpr-using-yolo/blob/main/seperation.py): `seperation.py`
- [Character detection](https://github.com/silvesterm05/alpr-using-yolo/blob/main/final.py): `final.py`
- **EfficientNet classifier**

### Exported data

- [Labeled character crops](https://github.com/silvesterm05/alpr-using-yolo/blob/main/seperation.py): `seperation.py`
- [Labeled crop export](https://github.com/silvesterm05/alpr-using-yolo/blob/main/final.py): `final.py`
- **Coordinate CSVs**
- **Plate string**

### Other components

- **Operator**
- **Vehicle API**

## Connections

- Operator → Filtered downloader: runs (evidence: `download.py`)
- Filtered downloader → Vehicle API: requests records and images (evidence: `download.py`)
- Filtered downloader → Downloaded images: writes images (evidence: `download.py`)
- Operator → Mall downloader: runs (evidence: `marinaimg.py`)
- Mall downloader → Vehicle API: requests records and images (evidence: `marinaimg.py`)
- Mall downloader → Downloaded images: writes images (evidence: `marinaimg.py`)
- Operator → Input images: provides images
- Input images → Plate localization: reads images (evidence: `seperation.py`)
- Plate localization → Plate crop and upscale: crops detected plate (evidence: `seperation.py`)
- Plate crop and upscale → Character detection: detects characters (evidence: `seperation.py`)
- Character detection → Labeled character crops: orders and exports crops (evidence: `seperation.py`)
- Labeled character crops → Coordinate CSVs: writes coordinates (evidence: `seperation.py`)
- Operator → Input images: provides images
- Input images → Character detection: reads and upscales (evidence: `final.py`)
- Character detection → Labeled crop export: exports filename-labeled crops (evidence: `final.py`)
- Labeled crop export → Coordinate CSVs: writes coordinates (evidence: `final.py`)
- EfficientNet classifier → Plate string: classifies characters (evidence: `README.md`)
