Object-Detection

💫 About Developer:

Banish J

🤖 AI & ML Developer | Data Science & Analytics Expert
🌐 Web Apps & IoT Skilled | Awesome UI Creator
💻 AI is my main focus! 👾

📞 Contact

☎️ 9444333914

📧 mail@banish.in

🌐 Banish

Core Concept

This script continuously captures video from the webcam, performs object detection on each frame using the MobileNet SSD model, and displays the results in real-time. It exits when the 'q' key is pressed and provides FPS information at the end.windows.

Importing Necessary Libraries

import numpy as np
import imutils
from imutils.video import FPS
import cv2
import time

• numpy: For numerical operations.
• imutils: A convenient library for basic image processing functions.
• imutils.video: Provides video processing utilities.
• cv2: The OpenCV library for computer vision tasks.
• time: A module to handle time-related tasks.

Setting Up Model and Parameters

prototxt = "MobileNetSSD_deploy.prototxt.txt"
model = "MobileNetSSD_deploy.caffemodel"
confThresh = 0.2

• prototxt: Path to the prototxt file containing the model architecture.
• model: Path to the pre-trained caffemodel file containing the learned weights.
• confThresh: Confidence threshold for object detection.

Class Labels and Colors

CLASSES = ["background", "aeroplane", "bicycle", "bird", "boat",
    "bottle", "bus", "car", "cat", "chair", "cow", "diningtable",
    "dog", "horse", "motorbike", "person", "pottedplant", "sheep",
    "sofa", "train", "tvmonitor"]
COLORS = np.random.uniform(0, 255, size=(len(CLASSES), 3))

• CLASSES: List of class labels for objects the model can detect.
• COLORS: Randomly generated colors for each class label.

Loading Model

net = cv2.dnn.readNetFromCaffe(prototxt, model)

• cv2.dnn.readNetFromCaffe: Loads the pre-trained model from the prototxt and caffemodel files.

Initializing Video Stream

vs = cv2.VideoCapture(0)
time.sleep(2.0)

• cv2.VideoCapture(0): Initializes the webcam for video capture. The argument 0 typically refers to the default webcam on the computer.
• time.sleep(2.0): Introduces a delay of 2 seconds to allow the webcam to warm up.

Initializing FPS Counter

fps = FPS().start()

• FPS(): Initializes the FPS counter.

Real-Time Object Detection Loop

while True:
    _, frame = vs.read()
    frame = imutils.resize(frame, width=1000)

    # Blob preprocessing...
    # Forward pass through the network...
    # Object detection and drawing...

• vs.read(): Captures a frame from the webcam.
• imutils.resize: Resizes the frame to have a maximum width of 1000 pixels.

Object Detection and Drawing

• The code performs object detection using the MobileNet SSD model on each frame captured from the webcam.
• Detected objects are drawn on the frame with bounding boxes and class labels.

Exiting the Loop

if key == ord("q"):
    break

• The loop exits if the 'q' key is pressed.

Cleanup

fps.stop()
vs.release()
cv2.destroyAllWindows()

• fps.stop(): Stops the FPS counter and displays the elapsed time and average FPS.
• vs.release(): Releases the webcam resource.
• cv2.destroyAllWindows(): Closes all OpenCV windows.