def check_for_anomalies(vc):
imagedump = []
for i in range(10):
rval, frame = vc.read()
frame = imresize(frame, (227, 227, 3))
#Convert the Image to Grayscale
gray = 0.2989 * frame[:, :,
0] + 0.5870 * frame[:, :,
1] + 0.1140 * frame[:, :, 2]
gray = (gray - gray.mean()) / gray.std()
gray = np.clip(gray, 0, 1)
imagedump.append(gray)
imagedump = np.array(imagedump)
imagedump.resize(227, 227, 10)
imagedump = np.expand_dims(imagedump, axis=0)
imagedump = np.expand_dims(imagedump, axis=4)
output = model.predict(imagedump)
loss = mean_squared_loss(imagedump, output)
if loss > threshold:
result = 'Anomalies Detected'
else:
result = 'No Anomalies'
return result
#Convert the Image to Grayscale
gray=0.2989*frame[:,:,0]+0.5870*frame[:,:,1]+0.1140*frame[:,:,2]
gray=(gray-gray.mean())/gray.std()
gray=np.clip(gray,0,1)
imagedump.append(gray)
imagedump=np.array(imagedump)
imagedump.resize(227,227,10)
imagedump=np.expand_dims(imagedump,axis=0)
imagedump=np.expand_dims(imagedump,axis=4)
print('Processing data')
output=model.predict(imagedump)
loss=mean_squared_loss(imagedump,output)
if loss>threshold:
print('Anomalies Detected')
frames = X_test.shape[2]
#Need to make number of frames divisible by 10
flag = 0 #Overall video flagq
frames = frames - frames % 10
X_test = X_test[:, :, :frames]
X_test = X_test.reshape(-1, 227, 227, 10)
X_test = np.expand_dims(X_test, axis=4)
for number, bunch in enumerate(X_test):
n_bunch = np.expand_dims(bunch, axis=0)
reconstructed_bunch = model.predict(n_bunch)
loss = mean_squared_loss(n_bunch, reconstructed_bunch)
if loss > threshold:
print("Anomalous bunch of frames at bunch number {}".format(number))
flag = 1
else:
print('Bunch Normal')
if flag == 1:
print("Anomalous Events detected")
from keras.preprocessing.image import img_to_array, load_img
import numpy as np
import glob
import os