######## Webcam Object Detection Using Tensorflow-trained Classifier #########

#

# Author: Evan Juras

# Date: 1/20/18

# Description:

# This program uses a TensorFlow-trained classifier to perform object detection.

# It loads the classifier uses it to perform object detection on a webcam feed.

# It draws boxes and scores around the objects of interest in each frame from

# the webcam.

## Some of the code is copied from Google's example at

## https://github.com/tensorflow/models/blob/master/research/object_detection/object_detection_tutorial.ipynb

## and some is copied from Dat Tran's example at

## https://github.com/datitran/object_detector_app/blob/master/object_detection_app.py

## but I changed it to make it more understandable to me.

# Import packages

import os

import cv2

import numpy as np

import tensorflow as tf

import sys

from PIL import Image

# This is needed since the notebook is stored in the object_detection folder.

sys.path.append("..")

# Import utilites

from utils import label_map_util

from utils import visualization_utils as vis_util

# Name of the directory containing the object detection module we're using

MODEL_NAME = 'inference_graph'

# Grab path to current working directory

CWD_PATH = os.getcwd()

# Path to frozen detection graph .pb file, which contains the model that is used

# for object detection.

PATH_TO_CKPT = os.path.join(CWD_PATH,MODEL_NAME,'frozen_inference_graph.pb')

#set PYTHONPATH=C:\tensorflow1\models;C:\tensorflow1\models\research;C:\tensorflow1\models\research\slim

# Path to label map file

PATH_TO_LABELS = os.path.join(CWD_PATH,'training','labelmap.pbtxt')

# Number of classes the object detector can identify

NUM_CLASSES = 5

## Load the label map.

# Label maps map indices to category names, so that when our convolution

# network predicts `5`, we know that this corresponds to `king`.

# Here we use internal utility functions, but anything that returns a

# dictionary mapping integers to appropriate string labels would be fine

label_map = label_map_util.load_labelmap(PATH_TO_LABELS)

categories = label_map_util.convert_label_map_to_categories(label_map, max_num_classes=NUM_CLASSES, use_display_name=True)

category_index = label_map_util.create_category_index(categories)

# Load the Tensorflow model into memory.

detection_graph = tf.Graph()

with detection_graph.as_default():

od_graph_def = tf.GraphDef()

with tf.gfile.GFile(PATH_TO_CKPT, 'rb') as fid:

serialized_graph = fid.read()

od_graph_def.ParseFromString(serialized_graph)

tf.import_graph_def(od_graph_def, name='')

sess = tf.Session(graph=detection_graph)

# Define input and output tensors (i.e. data) for the object detection classifier

# Input tensor is the image

image_tensor = detection_graph.get_tensor_by_name('image_tensor:0')

# Output tensors are the detection boxes, scores, and classes

# Each box represents a part of the image where a particular object was detected

detection_boxes = detection_graph.get_tensor_by_name('detection_boxes:0')

# Each score represents level of confidence for each of the objects.

# The score is shown on the result image, together with the class label.

detection_scores = detection_graph.get_tensor_by_name('detection_scores:0')

detection_classes = detection_graph.get_tensor_by_name('detection_classes:0')

# Number of objects detected

num_detections = detection_graph.get_tensor_by_name('num_detections:0')

# Initialize webcam feed

video = cv2.VideoCapture(0)

ret = video.set(3,1280)

ret = video.set(4,720)

img=Image.open('checkSize.jpg')

width, height = img.size

print(width)

print(height)

flagSide = 'left'

while(True):

# Acquire frame and expand frame dimensions to have shape: [1, None, None, 3]

# i.e. a single-column array, where each item in the column has the pixel RGB value

ret, frame = video.read()

frame_expanded = np.expand_dims(frame, axis=0)

# Perform the actual detection by running the model with the image as input

(boxes, scores, classes, num) = sess.run(

[detection_boxes, detection_scores, detection_classes, num_detections],

feed_dict={image_tensor: frame_expanded})

if len(classes[0] > 0):

listOf4Indexes = np.where(classes[0] == 4)

#print(listOf4Indexes)

# print(len(classes[0]))

# print(len(listOf4Indexes[0]))

# print('len(boxes)={} len(boxes[0])={} len(boxes[0][0]={}'.format(len(boxes),len(boxes[0]),len(boxes[0][0])))

c = np.zeros(shape=(1,len(classes[0])-len(listOf4Indexes[0])))

b = np.zeros((1,len(classes[0])-len(listOf4Indexes[0]),4))

s = np.zeros(shape=(1,len(classes[0])-len(listOf4Indexes[0])))

for FourIndex in listOf4Indexes:

# classes[0]=np.delete(classes[0],FourIndex

# boxes[0]=np.delete(boxes[0],FourIndex)

# scores[0]=np.delete(scores[0],FourIndex)

c[0]=np.delete(classes[0],FourIndex)

b[0]=np.delete(boxes,FourIndex,axis=1)

s[0]=np.delete(scores[0],FourIndex)

# print(len(c[0]))

# Draw the results of the detection (aka 'visulaize the results')

vis_util.visualize_boxes_and_labels_on_image_array(

frame,

np.squeeze(b),

np.squeeze(c).astype(np.int32),

np.squeeze(s),

category_index,

use_normalized_coordinates=True,

line_thickness=8,

min_score_thresh=0.60)

#print(boxes[0][1])

if classes[0][0] == 2 and scores[0][0] > 0.96:

print('I see a Wallet')

if width * boxes[0][0][1] <= width/2:

flagSide = 'left'

#imageName = 'left-'+imageName

#print('width/2={} boxes[0][1]={}'.format(width/2,boxes[0][1]))

else :

flagSide = 'right'

#print('Wallet is on {} side'.format(flagSide))

#imageName = 'right-'+imageName

#print('width/2={} boxes[0][1]={}'.format(width/2,boxes[0][1]))

print('Wallet is on {} side'.format(flagSide))

# All the results have been drawn on the frame, so it's time to display it.

cv2.namedWindow('Object detector', cv2.WINDOW_NORMAL)

cv2.resizeWindow('Object detector',800,600)

cv2.imshow('Object detector', frame)

# Press 'q' to quit

if cv2.waitKey(1) == ord('q'):

break

# Clean up

video.release()

cv2.destroyAllWindows()

#----------------------------------------------------------------------------------------------------------------------------------------------------------

MODEL_NAME = 'inferenceGraphs/inference_graphWithUnknown'

IMAGES_NAME = []

# Grab path to current working directory111

CWD_PATH = os.getcwd()

# Path to frozen detection graph .pb file, which contains the model that is used

# for object detection.

PATH_TO_CKPT = os.path.join(CWD_PATH,MODEL_NAME,'frozen_inference_graph.pb')

# Path to label map file

PATH_TO_LABELS = os.path.join(CWD_PATH,'training','labelmap.pbtxt')

#taking new images for registering an object------------------------------------------

cam = cv2.VideoCapture(0)

cv2.namedWindow('test')

img_counter = 0

while True:

ret,frame = cam.read()

cv2.imshow("test",frame)

if not ret:

break

k = cv2.waitKey(1)

if k%256 == 27:

print("Escape hit...")

break

elif k%256 == 32:

img_name = "Took_Image_{}.jpg".format(img_counter)

cv2.imwrite(img_name,frame)

print("{} written!".format(img_name))

img_counter += 1

cam.release()

cv2.destroyAllWindows()

import pdb; pdb.set_trace()

#-----------------------------------------------------------------------------------

#creating augmented images----------------------------------------------------------

types = ('newImagesForAugment\*jpg','newImagesForAugment\*jpeg')

for files in types:

IMAGES_NAME.extend(glob.glob(files))

# Path to image

sess1 = tf.Session()

for singleImage in IMAGES_NAME:

img = pt.imread(singleImage)

tf_img = tf.convert_to_tensor(img)

brght_img = tf.image.flip_left_right(tf_img)

fileToSave = tf.image.encode_jpeg(brght_img)

fname = tf.constant(singleImage.split('.')[0]+'output2.jpg')

fwrite = tf.write_file(fname,fileToSave)

result = sess1.run(fwrite)

brght_img = tf.image.transpose_image(tf_img)

fileToSave = tf.image.encode_jpeg(brght_img)

fname = tf.constant(singleImage.split('.')[0]+'transposed.jpg')

fwrite = tf.write_file(fname,fileToSave)

result = sess1.run(fwrite)

brght_img = tf.image.rot90(tf_img)

fileToSave = tf.image.encode_jpeg(brght_img)

fname = tf.constant(singleImage.split('.')[0]+'rot90.jpg')

fwrite = tf.write_file(fname,fileToSave)

result = sess1.run(fwrite)

brght_img = tf.image.rot90(tf_img,k=3)

fileToSave = tf.image.encode_jpeg(brght_img)

fname = tf.constant(singleImage.split('.')[0]+'rot90.jpg')

fwrite = tf.write_file(fname,fileToSave)

result = sess1.run(fwrite) #central_crop

brght_img = tf.image.central_crop(tf_img,0.70)

fileToSave = tf.image.encode_jpeg(brght_img)

fname = tf.constant(singleImage.split('.')[0]+'centralcropped.jpg')

fwrite = tf.write_file(fname,fileToSave)

result = sess1.run(fwrite)

# brght_img = tf.contrib.image.rotate(tf_img, 30 * math.pi / 180, interpolation='BILINEAR')

# fileToSave = tf.image.encode_jpeg(brght_img)

# fname = tf.constant(singleImage.split('.')[0]+'30degrees.jpg')

# fwrite = tf.write_file(fname,fileToSave)

# result = sess1.run(fwrite)

src_im = Image.open(singleImage)

rotated = scipy.ndimage.rotate(src_im, 20, cval=210)

scipy.misc.imsave(singleImage.split('.')[0]+'rotatedTest.jpg', rotated)

img = pt.imread(singleImage.split('.')[0]+'rotatedTest.jpg')

tf_img = tf.convert_to_tensor(img)

brght_img = tf.image.central_crop(tf_img,0.60)

fileToSave = tf.image.encode_jpeg(brght_img)

fname = tf.constant(singleImage.split('.')[0]+'rotatedTest.jpg')

fwrite = tf.write_file(fname,fileToSave)

result = sess1.run(fwrite)

# angle = 45

# size = 200, 200

# dst_im = Image.new("RGBA", (196,283), "blue" )

# im = src_im.convert('RGBA')

# rot = im.rotate( angle, expand=1 ).resize(size)

# dst_im.paste( rot, (0, 0), rot )

# dst_im = dst_im.convert('RGB')

# dst_im.save(singleImage.split('.')[0]+'45bllue.jpg','JPEG')

# Number of classes the object detector can identify

#--------------------------------------------------------------------------------------------

#Object detection----------------------------------------------------------------------------

import pdb; pdb.set_trace()

NUM_CLASSES = 4

IMAGES_NAME = []

types = ('newImagesForAugment\*jpg','newImagesForAugment\*jpeg')

for files in types:

IMAGES_NAME.extend(glob.glob(files))

# Load the label map.

# Label maps map indices to category names, so that when our convolution

# network predicts `5`, we know that this corresponds to `king`.

# Here we use internal utility functions, but anything that returns a

# dictionary mapping integers to appropriate string labels would be fine

label_map = label_map_util.load_labelmap(PATH_TO_LABELS)

categories = label_map_util.convert_label_map_to_categories(label_map, max_num_classes=NUM_CLASSES, use_display_name=True)

category_index = label_map_util.create_category_index(categories)

detection_graph = tf.Graph()

with detection_graph.as_default():

od_graph_def = tf.GraphDef()

with tf.gfile.GFile(PATH_TO_CKPT, 'rb') as fid:

serialized_graph = fid.read()

od_graph_def.ParseFromString(serialized_graph)

tf.import_graph_def(od_graph_def, name='')

sess = tf.Session(graph=detection_graph)

for singleImage in IMAGES_NAME:

PATH_TO_IMAGE = os.path.join(CWD_PATH,singleImage)

# Load the Tensorflow model into memory.

im = Image.open(PATH_TO_IMAGE)

width, height = im.size

print(width)

print(height)

# Define input and output tensors (i.e. data) for the object detection classifier

# Input tensor is the image

image_tensor = detection_graph.get_tensor_by_name('image_tensor:0')

# Output tensors are the detection boxes, scores, and classes

# Each box represents a part of the image where a particular object was detected

detection_boxes = detection_graph.get_tensor_by_name('detection_boxes:0')

# Each score represents level of confidence for each of the objects.

# The score is shown on the result image, together with the class label.

detection_scores = detection_graph.get_tensor_by_name('detection_scores:0')

detection_classes = detection_graph.get_tensor_by_name('detection_classes:0')

# Number of objects detected

num_detections = detection_graph.get_tensor_by_name('num_detections:0')

# Load image using OpenCV and

# expand image dimensions to have shape: [1, None, None, 3]

# i.e. a single-column array, where each item in the column has the pixel RGB value

image = cv2.imread(PATH_TO_IMAGE)

image_expanded = np.expand_dims(image, axis=0)

# Perform the actual detection by running the model with the image as input

(boxes, scores, classes, num) = sess.run(

[detection_boxes, detection_scores, detection_classes, num_detections],

feed_dict={image_tensor: image_expanded})

tempBoxes = np.copy(boxes)

# Draw the results of the detection (aka 'visulaize the results')

obj_count = 1

print(boxes)

print(boxes[0][0:obj_count][:])

boxes= boxes[0][0:obj_count][:]

#print(boxes[0])

i = 0

while i < len(boxes[0]):

if i % 2 == 0:

boxes[0][i] = (height * boxes[0][i]).astype(np.int32)

else:

boxes[0][i] = (width * boxes[0][i]).astype(np.int32)

i += 1

#print(classes)

#print(tempBoxes)

#print(singleImage)

#print(PATH_TO_IMAGE)

imageName = singleImage.split('\\')[1]

#print(imageName)

vis_util.visualize_boxes_and_labels_on_image_array(

image,

np.squeeze(tempBoxes),

np.squeeze(classes).astype(np.int32),

np.squeeze(scores),

category_index,

use_normalized_coordinates=True,

line_thickness=8,

min_score_thresh=0.50)

#--------------------------------------------------------------------------------------

if classes[0][0] == 2 and scores[0][0] > 0.96:

print('I see a Wallet')

if boxes[0][1] <= width/2:

print('Wallet is on left side')

imageName = 'left-'+imageName

print('width/2={} boxes[0][1]={}'.format(width/2,boxes[0][1]))

else :

print('Wallet is on right side')

imageName = 'right-'+imageName

print('width/2={} boxes[0][1]={}'.format(width/2,boxes[0][1]))

#Saving new xml annotated image--------------------------------------------------------

# All the results have been drawn on image. Now display the image.

#cv2.imshow('Object detector', image)

filepath = os.path.join(singleImage.split('\\')[0]+'\\annotatedImages\\',imageName)

print(filepath)

cv2.imwrite(filepath,image)

root = ET.Element("annotation")

folder = ET.SubElement(root, "folder")

folder.text = singleImage.split('\\')[0]

filename = ET.SubElement(root, "filename")

filename.text = imageName

path = ET.SubElement(root, "path")

path.text = PATH_TO_IMAGE

source = ET.SubElement(root, "source")

ET.SubElement(source, "database").text = "Unknown"

size = ET.SubElement(root, "size")

ET.SubElement(size, "width").text = str(width)

ET.SubElement(size, "height").text = str(height)

ET.SubElement(size, "depth").text = "3"

segmented = ET.SubElement(root, "segmented")

segmented.text = "0"

objectTag = ET.SubElement(root, "object")

name = ET.SubElement(objectTag,"name")

name.text = "tempClass1"

pose = ET.SubElement(objectTag,"pose")

pose.text = "Unspecified"

truncated = ET.SubElement(objectTag,"truncated")

truncated.text = "0"

difficult = ET.SubElement(objectTag,"difficult")

difficult.text = "0"

bndbox = ET.SubElement(objectTag,"bndbox")

ET.SubElement(bndbox, "xmin").text = str(boxes[0][1])

ET.SubElement(bndbox, "ymin").text = str(boxes[0][0])

ET.SubElement(bndbox, "xmax").text = str(boxes[0][3])

ET.SubElement(bndbox, "ymax").text = str(boxes[0][2])

tree = ET.ElementTree(root)

finalFileName = 'newImagesForAugment\\'+imageName.split('.')[0] + '.xml'

tree.write(finalFileName)

#time.sleep(2)

#---------------------------------------------------------------------------------------------

# Press any key to close the image

#cv2.waitKey(0)

# Clean up

#cv2.destroyAllWindows()

import pdb; pdb.set_trace()

#---------------------------------------------------------------------------------------------------------------------------------------------------------

MODEL_NAME = 'inference_graph'

# Grab path to current working directory

CWD_PATH = os.getcwd()

# Path to frozen detection graph .pb file, which contains the model that is used

# for object detection.

PATH_TO_CKPT = os.path.join(CWD_PATH,MODEL_NAME,'frozen_inference_graph.pb')

#set PYTHONPATH=C:\tensorflow1\models;C:\tensorflow1\models\research;C:\tensorflow1\models\research\slim

# Path to label map file

PATH_TO_LABELS = os.path.join(CWD_PATH,'training','labelmap.pbtxt')

# Number of classes the object detector can identify

NUM_CLASSES = 6

## Load the label map.

# Label maps map indices to category names, so that when our convolution

# network predicts `5`, we know that this corresponds to `king`.

# Here we use internal utility functions, but anything that returns a

# dictionary mapping integers to appropriate string labels would be fine

label_map = label_map_util.load_labelmap(PATH_TO_LABELS)

categories = label_map_util.convert_label_map_to_categories(label_map, max_num_classes=NUM_CLASSES, use_display_name=True)

category_index = label_map_util.create_category_index(categories)

# Load the Tensorflow model into memory.

detection_graph = tf.Graph()

with detection_graph.as_default():

od_graph_def = tf.GraphDef()

with tf.gfile.GFile(PATH_TO_CKPT, 'rb') as fid:

serialized_graph = fid.read()

od_graph_def.ParseFromString(serialized_graph)

tf.import_graph_def(od_graph_def, name='')

sess = tf.Session(graph=detection_graph)

# Define input and output tensors (i.e. data) for the object detection classifier

# Input tensor is the image

image_tensor = detection_graph.get_tensor_by_name('image_tensor:0')

# Output tensors are the detection boxes, scores, and classes

# Each box represents a part of the image where a particular object was detected

detection_boxes = detection_graph.get_tensor_by_name('detection_boxes:0')

# Each score represents level of confidence for each of the objects.

# The score is shown on the result image, together with the class label.

detection_scores = detection_graph.get_tensor_by_name('detection_scores:0')

detection_classes = detection_graph.get_tensor_by_name('detection_classes:0')

# Number of objects detected

num_detections = detection_graph.get_tensor_by_name('num_detections:0')

# Initialize webcam feed

video = cv2.VideoCapture(0)

ret = video.set(3,1280)

ret = video.set(4,720)

img=Image.open('checkSize.jpg')

width, height = img.size

print(width)

print(height)

flagSide = 'left'

while(True):

# Acquire frame and expand frame dimensions to have shape: [1, None, None, 3]

# i.e. a single-column array, where each item in the column has the pixel RGB value

ret, frame = video.read()

frame_expanded = np.expand_dims(frame, axis=0)

# Perform the actual detection by running the model with the image as input

(boxes, scores, classes, num) = sess.run(

[detection_boxes, detection_scores, detection_classes, num_detections],

feed_dict={image_tensor: frame_expanded})

if len(classes[0] > 0):

listOf4Indexes = np.where(classes[0] == 4)

#print(listOf4Indexes)

# print(len(classes[0]))

# print(len(listOf4Indexes[0]))

# print('len(boxes)={} len(boxes[0])={} len(boxes[0][0]={}'.format(len(boxes),len(boxes[0]),len(boxes[0][0])))

c = np.zeros(shape=(1,len(classes[0])-len(listOf4Indexes[0])))

b = np.zeros((1,len(classes[0])-len(listOf4Indexes[0]),4))

s = np.zeros(shape=(1,len(classes[0])-len(listOf4Indexes[0])))

for FourIndex in listOf4Indexes:

# classes[0]=np.delete(classes[0],FourIndex

# boxes[0]=np.delete(boxes[0],FourIndex)

# scores[0]=np.delete(scores[0],FourIndex)

c[0]=np.delete(classes[0],FourIndex)

b[0]=np.delete(boxes,FourIndex,axis=1)

s[0]=np.delete(scores[0],FourIndex)

# print(len(c[0]))

# Draw the results of the detection (aka 'visulaize the results')

vis_util.visualize_boxes_and_labels_on_image_array(

frame,

np.squeeze(b),

np.squeeze(c).astype(np.int32),

np.squeeze(s),

category_index,

use_normalized_coordinates=True,

line_thickness=8,

min_score_thresh=0.60)

#print(boxes[0][1])

if classes[0][0] == 2 and scores[0][0] > 0.96:

print('I see a Wallet')

if width * boxes[0][0][1] <= width/2:

flagSide = 'left'

#imageName = 'left-'+imageName

#print('width/2={} boxes[0][1]={}'.format(width/2,boxes[0][1]))

else :

flagSide = 'right'

#print('Wallet is on {} side'.format(flagSide))

#imageName = 'right-'+imageName

#print('width/2={} boxes[0][1]={}'.format(width/2,boxes[0][1]))

print('Wallet is on {} side'.format(flagSide))

# All the results have been drawn on the frame, so it's time to display it.

cv2.namedWindow('Object detector', cv2.WINDOW_NORMAL)

cv2.resizeWindow('Object detector',800,600)

cv2.imshow('Object detector', frame)

# Press 'q' to quit

if cv2.waitKey(1) == ord('q'):

break

# Clean up

video.release()

cv2.destroyAllWindows()