Final Output Overview

The output system was designed to make the project easier to understand visually. Instead of only showing raw data or CSV files, the dashboards connect images, extracted features, spectral trends, and model results in interactive views.

RealSense Dashboard

Shows daily stress detection outputs using RGB frames, segmentation overlays, confidence values, VARI trends, and canopy area trends.

Hyperspectral Tracker

Shows VOG1 and NDVI temporal trends from the Cubert hyperspectral workflow.

Visual Monitoring Outputs

Combines image outputs, feature summaries, trend charts, and model ready information for project presentation and future digital twin use.

Interactive RealSense Stress Detection Dashboard

This dashboard is the main visualization output from the RealSense workflow. It shows the original RGB image, segmentation overlay, daily plant metrics, model confidence, VARI trend, and canopy area trend.

RealSense Dashboard Output

The dashboard brings together image outputs, segmentation results, feature changes, and stress classification into one interactive monitoring view.

Interactive Hyperspectral Vegetation Index Tracker

This dashboard visualizes Lia’s hyperspectral temporal outputs using VOG1 and NDVI. It summarizes the latest values, index ranges, and daily trends from Set 2 Cubert hyperspectral data.

Hyperspectral Dashboard Output

The tracker shows how VOG1 and NDVI changed across the selected experiment dates using an interactive dashboard style layout.

Generated Outputs

The workflows produced visual and structured outputs that can be used for reporting, dashboard display, and future digital twin integration.

RGB Frames Segmentation Overlays CSV Feature Tables Daily Aggregates VARI Trends Canopy Area Trends VOG1 Trends NDVI Trends Model Confidence Dashboard Views

These outputs help turn raw RealSense and hyperspectral data into interpretable visual information for soybean growth monitoring and drought stress detection.

Why Visualization Matters

Visualization makes the pipeline easier to explain because it connects raw plant images, segmentation results, extracted features, hyperspectral trends, and model predictions. This helps users see both the final outputs and the visual evidence behind them.

← Back to Home