smarthome-presence-detect/camera/counter_people.py

#!/usr/bin/env python

import argparse
import numpy as np
import cv2
import imutils
from imutils.object_detection import non_max_suppression
from video_stream import imagezmq


'''
Usage:

python peopleCounter.py -i PATH_TO_IMAGE  # Reads and detect people in a single local stored image
python peopleCounter.py -c  # Attempts to detect people using webcam
'''

HOGCV = cv2.HOGDescriptor()
HOGCV.setSVMDetector(cv2.HOGDescriptor_getDefaultPeopleDetector())
#HOGCV.set
VERBOSITY = False

def detector(image):
    '''
    @image is a numpy array
    '''

    clone = image.copy()



    (rects, _) = HOGCV.detectMultiScale(image, winStride=(2, 2), padding=(8, 8), scale=1.05)

    # draw the original bounding boxes
    for (x, y, w, h) in rects:
        cv2.rectangle(clone, (x, y), (x + w, y + h), (0, 0, 255), 2)

    # Applies non-max supression from imutils package to kick-off overlapped
    # boxes
    rects = np.array([[x, y, x + w, y + h] for (x, y, w, h) in rects])
    result = non_max_suppression(rects, probs=None, overlapThresh=0.65)

    return result

def args_parser():
    ''' images, videos, remote or a local camera feed allowed
        verbose for added debugging'''
    ap = argparse.ArgumentParser()
    ap.add_argument("-i", "--image", default=None, 
                    help="path to image test file directory")
    ap.add_argument("-c", "--camera", action="store_true", default=False, 
                    help="Set as true if you wish to use the camera")
    ap.add_argument("-v", "--video", default=None, 
                    help="path to the video file")
    ap.add_argument("-r", "--remote", action="store_true", default=False, 
                    help="video comes from remote source via imagezmq")
    ap.add_argument("--verbose", action="store_true", default=False,
                    help="increase output verbosity")
    args = vars(ap.parse_args())

    if args["verbose"]:
        VERBOSITY = True

    return args

def usage():
    print("usage: counter_people.py [-h] [-i IMAGE] [-c] [-v] [-r REMOTE] [--verbose]")
    print()
    print("optional arguments:")
    print("  -h, --help            show this help message and exit")
    print("  -i IMAGE, --image IMAGE")
    print("                        path to image test file directory")
    print("  -c, --camera          Set as true if you wish to use the camera")
    print("  -v, --video           path to the video file")
    print("  -r REMOTE, --remote REMOTE")
    print("                        video comes from remote source via imagezmq")
    print("  --verbose             increase output verbosity")

def localDetect(image_path):
    result = []
    image = cv2.imread(image_path)
    image = imutils.resize(image, width=min(400, image.shape[1]))
    if len(image) <= 0:
        print("[ERROR] could not read local image")
        return result
    print("[INFO] Detecting people")
    result = detector(image)

    if VERBOSITY:
    # shows the result
        for (xA, yA, xB, yB) in result:
            cv2.rectangle(image, (xA, yA), (xB, yB), (0, 255, 0), 2)

        cv2.imshow("result", image)
        cv2.waitKey(0)
        cv2.destroyWindow("result")

        #cv2.imwrite("result.png", np.hstack((clone, image)))
    return result#(result, image)


def videoDetect(cap):
    while True:
        # Capture frame-by-frame
        _, frame = cap.read()

        if frame is None:
            break

        frame = imutils.resize(frame, width=min(400, frame.shape[1]))
        result = detector(frame.copy())

        # shows the result
        for (xA, yA, xB, yB) in result:
            cv2.rectangle(frame, (xA, yA), (xB, yB), (0, 255, 0), 2)
        
        if VERBOSITY:
            cv2.imshow('frame', frame)
            cv2.waitKey(0)

        #if time.time() - init >= sample_time:
        if result:
            print("{} people detected.".format(len(result)))
        #init = time.time()

        if cv2.waitKey(1) & 0xFF == ord('q'):
            break

    # When everything done, release the capture
    cap.release()
    cv2.destroyAllWindows()

def remoteDetect(image_hub):
    while True:
        rpi_name, frame = image_hub.recv_image()
        image_hub.send_reply(b'OK')

        frame = imutils.resize(frame, width=min(400, frame.shape[1]))
        result = detector(frame.copy())

        # shows the result
        for (xA, yA, xB, yB) in result:
            cv2.rectangle(frame, (xA, yA), (xB, yB), (0, 255, 0), 2)
        
        #if VERBOSITY:
        cv2.imshow('frame', frame)
        #cv2.waitKey(0)

        #if time.time() - init >= sample_time:
        if len(result):
            print("{} people detected.".format(len(result)))
        #init = time.time()

        if cv2.waitKey(1) & 0xFF == ord('q'):
            break
        
        #cv2.imshow(rpi_name, frame) # 1 window for each RPi
        #cv2.waitKey(1)
        
    


def detectPeople(args):
    image_path = args["image"]
    video_path = args["video"]
    camera = True if args["camera"] else False
    remote = True if args["remote"] else False

    # Routine to read local image
    if image_path is not None:
        print("[INFO] Image path provided, attempting to read image")
        (result, image) = localDetect(image_path)
        print(str(len(result)) + " People detected.")

    elif video_path is not None:
        print("[INFO] Video path provided, reading video")
        cap = cv2.VideoCapture(video_path)
        videoDetect(cap)

    # Routine to read images from webcam
    elif camera:
        print("[INFO] Reading images from local camera")
        cap = cv2.VideoCapture(0)
        videoDetect(cap)

    elif remote:
        print("[INFO] Reading images from remote stream")
        image_hub = imagezmq.ImageHub()
        remoteDetect(image_hub)

    else:
        usage()



def main():
    args = args_parser()
    detectPeople(args)


if __name__ == '__main__':
    main()