SensorView

A Flask/Dash-based web application for sensor data visualization and analysis, featuring multi-level caching, interactive controls, and a WebWorker-based buffering system for smooth performance on large datasets.

Screenshots

3D Visualization and Filtering

2D Visualization and Filtering

Statistical Visualization

Features

Core Visualization Modes

• 3D Scatter Plots – interactive 3D visualization with color mapping, filtering, and temporal decay effects
• 2D Scatter Plots – dual-panel left/right views with synchronized data filtering
• Heatmap – correlation analysis across numerical columns
• Histogram – distribution analysis with customizable binning
• Parallel Categories (parcats) – multi-dimensional categorical exploration
• Violin Plot – distribution and density visualization

Advanced Features

• Efficient Data Processing – WebWorker + IndexedDB buffering pipeline for smooth frame playback
• Multiple File Support – load and compare multiple datasets simultaneously
• Session Management – isolated server-side data sessions for concurrent users
• Frame Navigation & Playback – slider, play/stop controls, and configurable decay trails
• Dark / Light Mode – theme switching for different viewing preferences
• Configuration Persistence – JSON-based settings saved between sessions
• Test Case Management – organised directory-based data file selection

Dependencies

Python Modules

• dash, dash-bootstrap-components, dash-daq – web framework and interactive UI components
• polars, pandas, numpy, pyarrow – data manipulation and analysis
• diskcache – server-side FanoutCache for session and frame data
• orjson – high-performance JSON serialization for API responses
• kaleido – static image export for plots
• flaskwebgui – desktop application wrapper
• waitress – production WSGI server (optional)

Usage

Data Preparation

Save data as .pkl (pickle) or .csv files under the ./data directory. Each test case needs an info.json that describes the column mappings.

Directory Structure

./data/
├── Example/
│   ├── info.json          # Required: column specifications
│   ├── sensor_data.csv
│   └── sensor_data.pkl
└── Another_Test/
    ├── info.json
    └── dataset.pkl

Configuration File (info.json)

{
  "slider": "Frame",
  "x_3d": "Latitude",
  "y_3d": "Longitude",
  "z_3d": "Height",
  "x_ref": "Host_Latitude",
  "y_ref": "Host_Longitude",
  "z_ref": "None",
  "keys": {
    "Height":    { "description": "Height (m)",       "decimal": 2, "type": "numerical" },
    "Longitude": { "description": "Longitude (m)",    "decimal": 2, "type": "numerical" },
    "Latitude":  { "description": "Latitude (m)",     "decimal": 2, "type": "numerical" },
    "Time":      { "description": "Time (s)",         "decimal": 2, "type": "numerical" },
    "Sensor":    { "description": "Sensor",                         "type": "categorical" },
    "Frame":     { "description": "Frame",            "decimal": 0, "type": "numerical" }
  }
}

Key fields:

• slider – column used for the frame navigation slider
• x_3d, y_3d, z_3d – default 3D axis columns
• x_ref, y_ref, z_ref – reference point columns; set to "None" to disable a reference axis
• keys – per-column metadata: description, decimal places, and type ("numerical" or "categorical")

Running the Application

Desktop app (default):

python app.py

Launches a desktop window via FlaskWebGUI on port 8521.

Development mode – set DEBUG = True in app.py:

python app.py
# Open http://0.0.0.0:8050/

Production server – uncomment the Waitress lines in app.py:

from waitress import serve
serve(app.server, listen="*:8000")