# vedika-g/stampede architecture

> Architecture diagram of the GitHub repository vedika-g/stampede, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-01). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/vedika-g/stampede
- Repository: https://github.com/vedika-g/stampede

## Overview

This is an AI-assisted stampede early-warning system: video is monitored, crowd signals are estimated, risk is assessed, and results are surfaced to operators and optionally persisted through a backend API. The dashboard implementation excerpt shows video capture, YOLO inference, CSRNet density estimation, risk calculation, UI updates, and an HTTP submission; README documents Optical Flow and broader emergency-operations views. Some README-listed capabilities are not represented as separate sampled modules, so their implementation links are not inferred.

Main entry points: [`dashboard/app.py`](https://github.com/vedika-g/stampede/blob/main/dashboard/app.py) (Dashboard workflow), [`backend/main.py`](https://github.com/vedika-g/stampede/blob/main/backend/main.py) (Analysis API), [`backend/database.py`](https://github.com/vedika-g/stampede/blob/main/backend/database.py) (Database access), [`csrnet/csrnet_model.py`](https://github.com/vedika-g/stampede/blob/main/csrnet/csrnet_model.py) (CSRNet model)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_monitoring["Video monitoring"]
  node_video_source["Video source"]
  node_video_capture["Video capture<br/>[app.py]"]
end

subgraph group_analysis["Crowd analysis"]
  node_yolo["YOLO detection<br/>[app.py]"]
  node_tracking["Person tracking<br/>[app.py]"]
  node_csrnet_model["CSRNet model<br/>[csrnet_model.py]"]
  node_density_estimation["Density estimation<br/>[app.py]"]
  node_optical_flow["Optical Flow<br/>[app.py]"]
  node_heatmap["Crowd heatmap<br/>[app.py]"]
  node_risk_engine["Risk assessment<br/>[risk_engine.py]"]
end

subgraph group_operations["Operator dashboard"]
  node_dashboard_app["Dashboard workflow<br/>[app.py]"]
  node_dashboard_ui["Dashboard views<br/>[dashboard_ui.py]"]
  node_alerts["Risk alerts<br/>[dashboard_ui.py]"]
end

subgraph group_persistence["Analysis persistence"]
  node_backend_api["Analysis API<br/>[main.py]"]
  node_database_access["Database access<br/>[database.py]"]
end

node_operator(("Operator"))
node_mysql[("MySQL")]

node_operator -->|"selects monitoring"| node_dashboard_app
node_video_source -->|"provides frames"| node_video_capture
node_dashboard_app -->|"opens source"| node_video_capture
node_dashboard_app -->|"runs detection"| node_yolo
node_dashboard_app -->|"loads model"| node_csrnet_model
node_dashboard_app -->|"estimates density"| node_density_estimation
node_dashboard_app -->|"analyzes movement"| node_optical_flow
node_dashboard_app -->|"tracks people"| node_tracking
node_dashboard_app -->|"generates heatmap"| node_heatmap
node_dashboard_app -->|"calculates risk"| node_risk_engine
node_dashboard_app -->|"updates views"| node_dashboard_ui
node_dashboard_app -->|"presents warnings"| node_alerts
node_dashboard_app -->|"posts analysis"| node_backend_api
node_backend_api -->|"gets connection"| node_database_access
node_database_access -->|"connects to"| node_mysql

click node_video_capture "https://github.com/vedika-g/stampede/blob/main/dashboard/app.py"
click node_yolo "https://github.com/vedika-g/stampede/blob/main/dashboard/app.py"
click node_tracking "https://github.com/vedika-g/stampede/blob/main/dashboard/app.py"
click node_csrnet_model "https://github.com/vedika-g/stampede/blob/main/csrnet/csrnet_model.py"
click node_density_estimation "https://github.com/vedika-g/stampede/blob/main/dashboard/app.py"
click node_optical_flow "https://github.com/vedika-g/stampede/blob/main/dashboard/app.py"
click node_heatmap "https://github.com/vedika-g/stampede/blob/main/dashboard/app.py"
click node_risk_engine "https://github.com/vedika-g/stampede/blob/main/risk_engine.py"
click node_dashboard_app "https://github.com/vedika-g/stampede/blob/main/dashboard/app.py"
click node_dashboard_ui "https://github.com/vedika-g/stampede/blob/main/dashboard/dashboard_ui.py"
click node_alerts "https://github.com/vedika-g/stampede/blob/main/dashboard/dashboard_ui.py"
click node_backend_api "https://github.com/vedika-g/stampede/blob/main/backend/main.py"
click node_database_access "https://github.com/vedika-g/stampede/blob/main/backend/database.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_video_source,node_video_capture toneBlue
class node_yolo,node_tracking,node_csrnet_model,node_density_estimation,node_optical_flow,node_heatmap,node_risk_engine,node_mysql toneAmber
class node_dashboard_app,node_dashboard_ui,node_alerts toneMint
class node_backend_api,node_database_access toneRose
class node_operator toneIndigo
```

## Components

### Video monitoring

- **Video source**
- [Video capture](https://github.com/vedika-g/stampede/blob/main/dashboard/app.py): `dashboard/app.py`

### Crowd analysis

- [YOLO detection](https://github.com/vedika-g/stampede/blob/main/dashboard/app.py): `dashboard/app.py`
- [Person tracking](https://github.com/vedika-g/stampede/blob/main/dashboard/app.py): `dashboard/app.py`
- [CSRNet model](https://github.com/vedika-g/stampede/blob/main/csrnet/csrnet_model.py): `csrnet/csrnet_model.py`
- [Density estimation](https://github.com/vedika-g/stampede/blob/main/dashboard/app.py): `dashboard/app.py`
- [Optical Flow](https://github.com/vedika-g/stampede/blob/main/dashboard/app.py): `dashboard/app.py`
- [Crowd heatmap](https://github.com/vedika-g/stampede/blob/main/dashboard/app.py): `dashboard/app.py`
- [Risk assessment](https://github.com/vedika-g/stampede/blob/main/risk_engine.py): `risk_engine.py`

### Operator dashboard

- [Dashboard workflow](https://github.com/vedika-g/stampede/blob/main/dashboard/app.py): `dashboard/app.py`
- [Dashboard views](https://github.com/vedika-g/stampede/blob/main/dashboard/dashboard_ui.py): `dashboard/dashboard_ui.py`
- [Risk alerts](https://github.com/vedika-g/stampede/blob/main/dashboard/dashboard_ui.py): `dashboard/dashboard_ui.py`

### Analysis persistence

- [Analysis API](https://github.com/vedika-g/stampede/blob/main/backend/main.py): `backend/main.py`
- [Database access](https://github.com/vedika-g/stampede/blob/main/backend/database.py): `backend/database.py`

### Other components

- **Operator**
- **MySQL**

## Connections

- Operator → Dashboard workflow: selects monitoring (evidence: `README.md`)
- Video source → Video capture: provides frames (evidence: `dashboard/app.py`)
- Dashboard workflow → Video capture: opens source (evidence: `dashboard/app.py`)
- Dashboard workflow → YOLO detection: runs detection (evidence: `dashboard/app.py`)
- Dashboard workflow → CSRNet model: loads model (evidence: `dashboard/app.py`)
- Dashboard workflow → Density estimation: estimates density (evidence: `dashboard/app.py`)
- Dashboard workflow → Optical Flow: analyzes movement (evidence: `README.md`)
- Dashboard workflow → Person tracking: tracks people (evidence: `README.md`)
- Dashboard workflow → Crowd heatmap: generates heatmap (evidence: `README.md`)
- Dashboard workflow → Risk assessment: calculates risk (evidence: `dashboard/app.py`)
- Dashboard workflow → Dashboard views: updates views (evidence: `dashboard/app.py`)
- Dashboard workflow → Risk alerts: presents warnings (evidence: `dashboard/app.py`)
- Dashboard workflow → Analysis API: posts analysis (evidence: `dashboard/app.py`)
- Analysis API → Database access: gets connection (evidence: `backend/main.py`)
- Database access → MySQL: connects to (evidence: `backend/database.py`)
