rekog.py

import os
import cv2
from PIL import Image
import numpy as np
from halo import FaceRecognition
from model import get_input_shape
import helper
import time
from register import register

FPS = 30


def gstreamer_pipeline(
    capture_width=640,
    capture_height=480,
    display_width=640,
    display_height=480,
    framerate=FPS,
    flip_method=0,
):
    return (
        "nvarguscamerasrc ! "
        "video/x-raw(memory:NVMM), "
        "width=(int)%d, height=(int)%d, "
        "format=(string)NV12, framerate=(fraction)%d/1 ! "
        "nvvidconv flip-method=%d ! "
        "video/x-raw, width=(int)%d, height=(int)%d, format=(string)BGRx ! "
        "videoconvert ! "
        "video/x-raw, format=(string)BGR ! appsink"
        % (
            capture_width,
            capture_height,
            framerate,
            flip_method,
            display_width,
            display_height,
        )
    )


def rekog():
    # n -> freeze frames for n seconds
    # f -> current frame
    # R -> wait time
    # r -> radius
    n = 30
    f = 0
    R = 0
    r = 180
    minWidthFace = 200
    isFreeze = False

    # detector = MTCNN()
    FR = FaceRecognition()

    # Using webcam
    video = cv2.VideoCapture(0)
    # Using gstreamer
    # video = cv2.VideoCapture(gstreamer_pipeline(), cv2.CAP_GSTREAMER)

    if not video.isOpened():
        raise Exception("Error opening the camera")

    # Open the camera and get the camera size
    ret, temp = video.read()

    # convert image to grayscale image
    gray_image = cv2.cvtColor(temp, cv2.COLOR_BGR2GRAY)
    # convert the grayscale image to binary image
    ret, thresh = cv2.threshold(gray_image, 127, 255, 0)
    # calculate moments of binary image
    M = cv2.moments(thresh)
    # calculate x,y coordinate of center
    CX = int(M["m10"] / M["m00"])
    CY = int(M["m01"] / M["m00"])

    while True:
        # Open camera for recognizing
        ret, frame = video.read()
        rekog_frame = frame[CY - r : CY + r, CX - r : CX + r]

        # Draw rekog circle
        cv2.circle(frame, center=(CX, CY), radius=r, color=(243, 144, 29), thickness=3)

        # When freeze don't analyze faces for 11 frames
        if isFreeze == True and 0 <= R <= 10:
            faces = []
            R += 1
            print(f"R = {R}")
        else:
            faces = helper.detectFacesLive(rekog_frame)
            R = 0
            isFreeze = False

        if len(faces) != 0 and faces[0][2] >= minWidthFace:
            x, y, w, h = faces[0]

            # Draw rectangle in face
            cv2.rectangle(rekog_frame, (x, y), (x + w, y + h), (0, 155, 255), 2)

            # Predict
            img = helper.detectFace(
                rekog_frame[y : y + h, x : x + w], get_input_shape(), stream=True
            )
            if img.shape[1:3] == get_input_shape():
                pred, score = FR.predict(img)

            # Count until freeze
            f += 1
            print(f"frame {f}")

            # Freeze to show the predicted face
            if f == n or score != "":
                isFreeze = True
                f = 0
                freeze_img = rekog_frame.copy()
                # Draw label class prediction
                cv2.rectangle(
                    freeze_img, (x, y + h + 65), (x + w, y + h), (0, 0, 0), cv2.FILLED
                )
                if pred == "":
                    cv2.rectangle(
                        freeze_img, (x, y), (x + w, y + h), (0, 155, 255), cv2.FILLED
                    )
                    cv2.putText(
                        freeze_img,
                        "Not Employee!",
                        (x + 3, y + h + 25),
                        cv2.FONT_HERSHEY_DUPLEX,
                        0.8,
                        (255, 255, 255),
                        1,
                    )
                elif pred != "":
                    cv2.putText(
                        freeze_img,
                        f"{pred}",
                        (x + 3, y + h + 25),
                        cv2.FONT_HERSHEY_DUPLEX,
                        0.8,
                        (255, 255, 255),
                        1,
                    )
                    cv2.putText(
                        freeze_img,
                        f"{score:.2f}",
                        (x + 3, y + h + 55),
                        cv2.FONT_HERSHEY_DUPLEX,
                        0.6,
                        (255, 255, 255),
                        1,
                    )
                else:
                    None
                cv2.imshow(f"Hello {pred}", freeze_img)
                cv2.waitKey(3000)
                cv2.destroyWindow(f"Hello {pred}")

        cv2.imshow("Hello Welcome to iNews Tower", frame)

        key = cv2.waitKey(1)
        if key == ord("q"):
            break
        elif key == ord("r"):
            video.release()
            cv2.destroyAllWindows()
            name = input("Enter your name: ")
            register(name=name)
            rekog()

    video.release()
    cv2.destroyAllWindows()