PROJECT OBJECTIVE-- To create a robust object detection system using OpenCV with the MobileNet-SSD model for real time recognition of objects from the COCO dataset, employing a camera-based video feed for dynamic visualization.
Description-- This project integrates a pre-trained MobileNet-SSD model to detect and classify objects in real time. The system uses COCO dataset classes and displays detected objects with bounding boxes and confidence levels.
Project Details-- Real-Time Detection:Achieve real-time object detection in video streams or camera feeds with high accuracy and speed.
Multi-Class Object Recognition: Detect and classify multiple objects of various classes simultaneously within a single frame, such as persons, vehicles, animals, and everyday items.
Integration with OpenCV: Utilize OpenCV for pre-processing (resizing,scaling, and normalization) and post-processing of images to enhance the performance and user experience.
User-Friendly Interface: Develop an interface for easy visualization and interaction, allowing users to see the detected objects and obtain relevant information in real-time.
Overview of Object Detection:-
Object Detection is a computer vision technique that involves identifying and locating objects within an image or video. Unlike image classification, which assigns a label to an entire image, object detection provides: Class Identification Localization
Why Choose YOLO?:-
If your project requires real-time object detection with good accuracy and the ability to handle multiple objects simultaneously, YOLO is the ideal choice. Its speed and simplicity make it suitable for a variety of applications, even on devices with limited processing power.
PROJECT MODULES Implementation Details :
- Dataset and Model Initialization: Load class labels from coco.names to map object IDs to names.Configure the model using the provided .pbtxt and .pb files.
- Camera and Video Stream Setup: Initialize the video feed (cv2.VideoCapture(0)). Set resolution and other parameters like brightness for optimal performance.
- Object Detection: Process each video frame using the cv2.dnn_DetectionModel. Perform object detection with a confidence threshold of 50%.
- Visualization and Annotation: Draw bounding boxes and display object names with confidence levels. Real-time visualization using cv2.imshow.
- User Control: Provide an interface to exit the program by pressing 'q'.
APPLICATION OF PROJECT
-
Surveillance Systems: Identifying and tracking people or vehicles.
-
Autonomous Vehicles: Detecting pedestrians, cars, traffic signs, etc.
-
Medical Imaging: Detecting tumors or abnormalities in scans.
How to run the code : 1. First open cmd . 2. Then select Project Directory using cd 3. For run the project use command : python project_name.py Here we use command for own project : python main.py
