-
Notifications
You must be signed in to change notification settings - Fork 0
/
GUI_REC.py
172 lines (142 loc) · 5.88 KB
/
GUI_REC.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
import sys
from PyQt5.QtGui import QImage, QPixmap, QFont
from PyQt5.QtCore import QTimer
from PyQt5.QtWidgets import QApplication, QDialog
from PyQt5.uic import loadUi
# from rec_web import *
import numpy as np
import cv2
import pickle
import numpy as np
# import tensorflow as tf
# from cnn_tf import cnn_model_fn
import os
import sqlite3
from keras.models import load_model
def get_image_size():
img = cv2.imread('gestures/0/100.jpg', 0)
return img.shape
def keras_process_image(img):
img = cv2.resize(img, (image_x, image_y))
img = np.array(img, dtype=np.float32)
img = np.reshape(img, (1, image_x, image_y, 1))
return img
def keras_predict(model, image):
processed = keras_process_image(image)
pred_probab = model.predict(processed)[0]
pred_class = list(pred_probab).index(max(pred_probab))
return max(pred_probab), pred_class
def get_pred_text_from_db(pred_class):
conn = sqlite3.connect("gesture_db.db")
cmd = "SELECT g_name FROM gesture WHERE g_id=" + str(pred_class)
cursor = conn.execute(cmd)
for row in cursor:
return row[0]
def get_hand_hist():
with open("hist", "rb") as f:
hist = pickle.load(f)
return hist
x, y, w, h = 300, 100, 300, 300
model = load_model('cnn_model_keras.h5')
image_x, image_y = get_image_size()
class MainWindow(QDialog):
def __init__(self):
super(MainWindow, self).__init__()
loadUi('OpenCv2.ui', self)
self.image = None
self.thresh = None
self.imgCrop = None
self.imgHSV = None
self.capture = None
self.font = QFont()
self.font.setFamily("Arial")
self.font.setPointSize(35)
self.myText.setFont(self.font)
self.stop_button.setEnabled(False)
self.start_button.clicked.connect(self.start_webcam)
self.stop_button.clicked.connect(self.stop_webcam)
def start_webcam(self):
# Capture video from external camera
self.capture = cv2.VideoCapture(1)
if not self.capture.read()[0]:
# if external camera not found, use inbuilt camera
self.capture = cv2.VideoCapture(0)
x, y, w, h = 300, 100, 300, 300
self.capture.set(cv2.CAP_PROP_FRAME_HEIGHT, 360)
self.capture.set(cv2.CAP_PROP_FRAME_WIDTH, 480)
self.stop_button.setEnabled(True)
self.timer = QTimer(self)
self.timer.timeout.connect(self.update_frame)
self.timer.start(1)
def update_frame(self):
text = " "
ret, self.image = self.capture.read()
self.image = cv2.flip(self.image, 1)
self.image = cv2.resize(self.image, (640, 480))
cv2.rectangle(self.image, (x, y), (x + w, y + h), (0, 255, 0), 2)
self.imgCrop = self.image[y:y + h, x:x + w]
self.imgHSV = cv2.cvtColor(self.imgCrop, cv2.COLOR_BGR2HSV)
hist = get_hand_hist()
dst = cv2.calcBackProject([self.imgHSV], [0, 1], hist, [0, 180, 0, 256], 1)
disc = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (10, 10))
cv2.filter2D(dst, -1, disc, dst)
blur = cv2.GaussianBlur(dst, (11, 11), 0)
blur = cv2.medianBlur(blur, 15)
thresh = cv2.threshold(blur, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)[1]
thresh = cv2.merge((thresh, thresh, thresh))
thresh = cv2.cvtColor(thresh, cv2.COLOR_BGR2GRAY)
self.thresh = thresh[y:y + h, x:x + w]
contours = cv2.findContours(thresh.copy(), cv2.RETR_TREE, cv2.CHAIN_APPROX_NONE)[0]
if len(contours) > 0:
contour = max(contours, key=cv2.contourArea)
# print(cv2.contourArea(contour))
if cv2.contourArea(contour) > 10000:
x1, y1, w1, h1 = cv2.boundingRect(contour)
save_img = thresh[y1:y1 + h1, x1:x1 + w1]
if w1 > h1:
save_img = cv2.copyMakeBorder(save_img, int((w1 - h1) / 2), int((w1 - h1) / 2), 0, 0,
cv2.BORDER_CONSTANT, (0, 0, 0))
elif h1 > w1:
save_img = cv2.copyMakeBorder(save_img, 0, 0, int((h1 - w1) / 2), int((h1 - w1) / 2),
cv2.BORDER_CONSTANT, (0, 0, 0))
pred_probab, pred_class = keras_predict(model, save_img)
if pred_probab * 100 > 80:
text = " "
text = get_pred_text_from_db(pred_class)
print(text)
self.myText.setText(text)
self.displayImage(self.image)
self.displayThresh(thresh)
def displayImage(self, img):
qformat = QImage.Format_Indexed8
if len(img.shape) == 3: # [0]=rows, [1]=cols, [2]=channels
if img.shape[2] == 4:
qformat = QImage.Format_RGBA8888
else:
qformat = QImage.Format_RGB888
outImage = QImage(img, img.shape[1], img.shape[0], img.strides[0], qformat)
# BGR to RGB
outImage = outImage.rgbSwapped()
self.camera_display.setPixmap(QPixmap.fromImage(outImage))
self.camera_display.setScaledContents(True)
def displayThresh(self, img):
qformat = QImage.Format_Indexed8
if len(img.shape) == 3: # [0]=rows, [1]=cols, [2]=channels
if img.shape[2] == 4:
qformat = QImage.Format_RGBA8888
else:
qformat = QImage.Format_RGB888
outImage = QImage(img, img.shape[1], img.shape[0], img.strides[0], qformat)
# BGR to RGB
outImage = outImage.rgbSwapped()
self.threshhold_display.setPixmap(QPixmap.fromImage(outImage))
self.threshhold_display.setScaledContents(True)
def stop_webcam(self):
self.capture.release()
self.timer.stop()
if __name__ == '__main__':
app = QApplication(sys.argv)
window = MainWindow()
window.setWindowTitle('ASL recognizer - Real Time Recognition')
window.show()
sys.exit(app.exec_())