HumanRecognition.py

import cv2
import numpy
import numpy as np
import matplotlib.pyplot as plt
import datetime
import face_recognition
import dlib
import pymongo

try:
    client = pymongo.MongoClient(host='hostname', port=27017, username='root', password='pass', authSource="admin")
    print("okay connection!!")
    db = client["users_db"]
    col = db.webcam_recognize
    transaction_db = db.transaction_table
except:
    print("error in connection")

human_cascade=cv2.CascadeClassifier('haarcascade_fullbody.xml')
class HumanRecognition():

    def __init__(self):
        pass

    def detect_human(self,image,value):
        human_image=image
        human_rectangle=human_cascade.detectMultiScale(human_image,scaleFactor=1.02,minNeighbors=4)
        if type(human_rectangle) == numpy.ndarray and type(human_rectangle) == numpy.ndarray:
            value=True
        for (x,y,w,h) in human_rectangle:
            font = cv2.FONT_HERSHEY_SIMPLEX
            image=cv2.putText(human_image, text='Person', org=((x,y+h)), fontFace=font, thickness=2,color=(0, 0, 255), lineType=cv2.LINE_AA, fontScale=1.25)
        return image,value

    def face_validation(self,frame):
        # Convert the image from BGR color (which OpenCV uses) to RGB color (which face_recognition uses)
        rgb_frame = frame[:, :, ::-1]
        facesCurFrame = face_recognition.face_locations(rgb_frame)
        encodesCurFrame = face_recognition.face_encodings(
            rgb_frame, facesCurFrame)
        for encodeFace, faceLoc in zip(encodesCurFrame, facesCurFrame):
            name = ''
            id = 0
            for record in col.find({}):
                flag = 0
                for k, v in record.items():
                    if (k.lower() == "pixelvalue"):
                        encodeListKnown = tuple(v[0])
                        matches = face_recognition.compare_faces(encodeListKnown, encodeFace)
                        faceDis = face_recognition.face_distance(encodeListKnown, encodeFace)
                        matchIndex = np.argmin(faceDis)
                        if matches[matchIndex]:
                            name = name.upper()
                            flag = 1
                            return ("Access Granted")
            if (flag == 0):
                return ("Access Denied")

    def video_cam(self):
        capture=cv2.VideoCapture(0)
        count = 1
        while True:
            capture.set(cv2.CAP_PROP_FPS, 2)
            ret,frame = capture.read()
            grayFrame = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
            if ret:
                value = None
                detection,val=self.detect_human(grayFrame,value)
                if val == True:
                    font = cv2.FONT_HERSHEY_SIMPLEX
                    validate_face=self.face_validation(frame)
                    if validate_face == "Access Granted":
                        if count == 1:
                            now = datetime.datetime.now()
                            now = now.strftime("%H:%M:%S")
                            cv2.putText(detection, text=validate_face, org=(35, 425), fontFace=font, thickness=2,
                                        color=(255, 0, 0), lineType=cv2.LINE_AA, fontScale=1.25)
                            cv2.imwrite(now + 'frame.jpeg', detection)
                            cv2.imshow('frame', detection)
                            count=count+1
                        if count > 1:
                            cv2.putText(detection, text=validate_face, org=(35, 425), fontFace=font, thickness=2,
                                        color=(255, 0, 0), lineType=cv2.LINE_AA, fontScale=1.25)
                            cv2.imshow('frame',detection)
                            count=count+1
                    if validate_face == "Access Denied":
                        if count == 1:
                            now = datetime.datetime.now()
                            now = now.strftime("%H:%M:%S")
                            cv2.putText(detection, text=validate_face, org=(35, 425), fontFace=font, thickness=2,
                                        color=(255, 0, 0), lineType=cv2.LINE_AA, fontScale=1.25)
                            cv2.imshow('frame', detection)
                            count=count+1
                        if count> 1 :
                            cv2.putText(detection, text=validate_face, org=(35, 425), fontFace=font, thickness=2,
                                        color=(255, 0, 0), lineType=cv2.LINE_AA, fontScale=1.25)
                            cv2.imshow('frame', detection)
                            count=count + 1
            if cv2.waitKey(1) & 0xFF == ord('q'):
                print("Camera Not Working")
                break

        capture.release()
        cv2.destroyAllWindows()

if __name__ == '__main__':
    initiate=HumanRecognition()
    initiate.video_cam()