https://github.com/konynour/tf-object-detection
This project implements an object detection system using TensorFlow and OpenCV, designed to identify and classify objects in images or video streams. By utilizing pre-trained models from the TensorFlow Object Detection API, this project allows users to detect multiple objects in real-time or from static images.
https://github.com/konynour/tf-object-detection
frozen jupyter-notebook matplotlip opencv python-script python3 tensorflow tensorflow-object-detection-api
Last synced: 8 months ago
JSON representation
This project implements an object detection system using TensorFlow and OpenCV, designed to identify and classify objects in images or video streams. By utilizing pre-trained models from the TensorFlow Object Detection API, this project allows users to detect multiple objects in real-time or from static images.
- Host: GitHub
- URL: https://github.com/konynour/tf-object-detection
- Owner: konynour
- Created: 2025-02-18T04:44:18.000Z (8 months ago)
- Default Branch: main
- Last Pushed: 2025-02-18T06:12:15.000Z (8 months ago)
- Last Synced: 2025-02-18T06:28:53.623Z (8 months ago)
- Topics: frozen, jupyter-notebook, matplotlip, opencv, python-script, python3, tensorflow, tensorflow-object-detection-api
- Language: Jupyter Notebook
- Homepage:
- Size: 64.8 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Object Detection using TensorFlow and OpenCV
## 📌 Overview
This project implements object detection using a pre-trained **SSD MobileNet v2** model from the TensorFlow Model Zoo. The model is loaded using OpenCV's Deep Neural Network (DNN) module, and it performs real-time object detection on images.
## ⚠️ Disclaimer
**This project is intended for educational purposes only. The author is not responsible for any misuse.**
## 🚀 Features
- Loads and processes images for object detection.
- Uses a pre-trained **SSD MobileNet v2** model.
- Detects multiple objects in an image and labels them with bounding boxes.
- Configurable detection threshold.
## 📂 Requirements
- Python 3.x
- OpenCV (`cv2`)
- Matplotlib
- numpy
- os
- urllib
- zipfile
- TensorFlow Model Zoo (pre-trained models)
## 🔧 Installation
1. Clone this repository:
```sh
git remote add origin https://github.com/konynour/TF-Object-Detection.git ```
2. Navigate to the project folder:
```sh
cd ObjectDetection
```
3. Install dependencies:
```sh
pip install opencv-python numpy matplotlib
```
## ⚙️ Usage
### 1️⃣ Load and Configure the Model
- Download and extract the **SSD MobileNet v2** model from TensorFlow Model Zoo.
- Place the `frozen_inference_graph.pb` and its configuration file `.pbtxt` inside the `models/` directory.
### 2️⃣ Run Object Detection on an Image
```python
import cv2
import os
from detect import detect_objects, display_objects
# Load an image
im = cv2.imread(os.path.join("images", "street.jpg"))
# Detect objects
objects = detect_objects(net, im)
# Display results
display_objects(im, objects)
```
### 3️⃣ Adjust Detection Threshold
If you want to adjust the confidence threshold, you can modify the detection function call:
```python
im = cv2.imread(os.path.join("images", "baseball.jpg"))
objects = detect_objects(net, im)
display_objects(im, objects, threshold=0.2)
```