Underwater-Resourse-Detection

Python + OpenCV GUI for underwater resource detection using Haar Cascade and CLAHE.

Underwater Resource Detection Using Image Processing

1. Overview

This project implements a Python-based underwater resource detection system with a Tkinter GUI, designed to identify and enhance underwater images and live video streams.
It combines Haar Cascade Classifier for object detection with Contrast Limited Adaptive Histogram Equalization (CLAHE) for image contrast enhancement, enabling more accurate recognition of marine fauna and flora in challenging underwater environments.

2. Features

• Real-time detection of underwater flora and fauna.
• Image enhancement using CLAHE to handle low-contrast conditions.
• Python Tkinter GUI for an easy-to-use interface.
• Distance and size estimation for detected objects.
• Works with both static images and live camera feeds.

3. Methodology

a. Dataset Preparation

• Collected underwater images containing flora and fauna.
• Cropped and labeled images into positive and negative datasets.
• Applied CLAHE to improve image contrast.

b. Training

• Used Haar Cascade Classifier trained on the processed dataset.

c. GUI Development

• Built in Python using Tkinter for user-friendly interaction.
• Allows selection of images or live video.
• Displays processed frames with detected objects highlighted.

d. Detection Process

• Runs detection in real-time.
• Calculates object distance, size, and coverage area.
• Allows parameter adjustments (e.g., CLAHE settings).

4. Results

Testing with annotated underwater fauna images achieved 80% detection accuracy.
The GUI displays real-time bounding boxes around detected objects, functioning effectively under various lighting conditions, including dark underwater environments

pip install -r requirements.txt