world best vehicle counting & detection with source code || all types of vehicle

 

world best vehicle counting & detection with source code  || all types of vehicle 

Introduction

The intelligent transport system (ITS) manages the transportation system and traffic system. The objectives of the intelligent transport system s to explored the transportation of the vehicle system as well as their features like vehicle detection and counting. Vehicle counting and classification can give accurate information about the vehicle density on a particular city road or highway which helps in advanced traffic monitoring and management. Various image processing techniques like frame differentiation, segmentation, filtering, morphological operations, blob detection, etc. are used using OpenCV software. The final count of the number of vehicles passed through the path of choice will be displayed and classified throughout the day. This is an advancement over the hardware counterparts which are made using sensors since sensors are severely affected by weather conditions such as temperature, lightning.

Source code :

• Install all the required libraryies 

like open cv , Numpy 

                     

import cv2

import numpy as np

from time import sleep

largura_min = 80

altura_min = 80

offset = 6

pos_linha = 550

# FPS to vídeo

delay = 60

detec = []

carros = 0

def pega_centro(x, y, w, h):

    x1 = int(w / 2)

    y1 = int(h / 2)

    cx = x + x1

    cy = y + y1

    return cx, cy

# video source input

cap = cv2.VideoCapture('video.mp4')

# OpenCV provides Background Subtraction algorithms.

# we can perform background Subtraction using matrix subtraction, i.e, just subtracting the static frame from the video.

# But this has a lot of drawbacks.

# It is a very less efficient algorithm for Background subtraction because it does not update itself.

# This problem is being handled by the Background Subtraction algorithms provided by OpenCV.

subtracao = cv2.bgsegm.createBackgroundSubtractorMOG()

while True:

    ret, frame1 = cap.read()

    tempo = float(1/delay)

    sleep(tempo)

    grey = cv2.cvtColor(frame1, cv2.COLOR_BGR2GRAY)

    blur = cv2.GaussianBlur(grey, (3, 3), 5)

    img_sub = subtracao.apply(blur)

    dilat = cv2.dilate(img_sub, np.ones((5, 5)))

    kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (5, 5))

    # The morphologyEx() of the method of the class Imgproc accepts src, dst, op, kernel as parameters

    dilatada = cv2.morphologyEx(dilat, cv2. MORPH_CLOSE, kernel)

    dilatada = cv2.morphologyEx(dilatada, cv2. MORPH_CLOSE, kernel)

    # OpenCV has findContour() function that helps in extracting the contours from the image.

    # It works best on binary images, so we should first apply thresholding techniques, Sobel edges, etc.

    contorno, h = cv2.findContours(dilatada, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)

    # it will create a line

    # Parameters:

    # image: It is the image on which line is to be drawn.

    # start_point: It is the starting coordinates of line.

    # end_point: It is the ending coordinates of line.

    # color: It is the color of line to be drawn.

    # thickness: It is the thickness of the line in px.

    cv2.line(frame1, (25, pos_linha), (1200, pos_linha), (176, 130, 39), 2)

    for(i, c) in enumerate(contorno):

        (x, y, w, h) = cv2.boundingRect(c)

        validar_contorno = (w >= largura_min) and (h >= altura_min)

        if not validar_contorno:

            continue

        cv2.rectangle(frame1, (x, y), (x+w, y+h), (0, 255, 0), 2)

        centro = pega_centro(x, y, w, h)

        detec.append(centro)

        cv2.circle(frame1, centro, 4, (0, 0, 255), -1)

        for (x, y) in detec:

            if (y < (pos_linha + offset)) and (y > (pos_linha-offset)):

                carros += 1

                cv2.line(frame1, (25, pos_linha), (1200, pos_linha), (0, 127, 255), 3)

                detec.remove((x, y))

                print("No. of cars detected : " + str(carros))

    # cv2.putText() method is used to draw a text string on any image.

    # Parameters: image, text, org(coordinate), font, color, thickness

    cv2.putText(frame1, "VEHICLE COUNT : "+str(carros), (320, 70), cv2.FONT_HERSHEY_COMPLEX, 2, (0, 0, 255), 4)

    cv2.imshow("Video Original", frame1)

    cv2.imshow(" Detectar ", dilatada)

    # To display the image, you can use the imshow() method of cv2

    if cv2.waitKey(1) == 27:

        break

cv2.destroyAllWindows()

cap.release()