def predict_object(msg, cam, model, ja_labels):
img_shape = (224, 224)
img_path = "/tmp/cap.jpg"
save_captured_image(img_path, cam)
img = image.load_img(img_path, target_size=img_shape)
x = image.img_to_array(img)
x = np.expand_dims(x, axis=0)
x = preprocess_input(x)
# predict
preds = model.predict(x)
# decode
results = decode_predictions(preds, top=1)[0]
for r in results:
print('Predicted:', r)
en_wnid = r[0]
acc = r[2]
# translate english label to japanese one.
ja_label = en2ja_label(ja_labels, en_wnid, acc)
print(ja_label)
if ja_label != "":
call_jtalk(ja_label)
def predict(self, frame):
# expand 3D RGB frame into 4D "batch"
sample = np.expand_dims(frame, axis=0)
processed_sample = preprocess_input(sample.astype(np.float32))
features = self.conv_base.predict(processed_sample)
decoded_features = decode_predictions(features)
return decoded_features
def decodepred(network_name, preds):
if (network_name == "ResNet50"):
preds = resnet50.decode_predictions(preds, top=3)[0]
elif (network_name == "MobileNetV2"):
preds = mobilenetv2.decode_predictions(preds, top=3)[0]
elif (network_name == "VGG19"):
preds = vgg19.decode_predictions(preds, top=3)[0]
elif (network_name == "SqueezeNet"):
preds = imagenet_utils.decode_predictions(preds, top=3)[0]
return x
def result(image):
K.clear_session()
#model = MobileNetV2(weights="imagenet", include_top=True)
#model = MobileNet(weights="imagenet", include_top=True)
model = DenseNet121(weights="imagenet", include_top=True)
x = np.expand_dims(image, axis=0)
x = image.reshape(x.shape[0], 224, 224, 3)
preds = model.predict(x)
result = decode_predictions(preds, top=3)[0]
result_str = ""
for _, name, score in result:
result_str += "{}: {:.2%} ".format(name, score)
return result_str
# Lets optimize the model for the Jetson's GPU
input_model = load_model(model_input_path)
frozenmodel = FrozenGraph(input_model, (224, 224, 3))
print('FrozenGraph build.')
model = TftrtEngine(frozenmodel, 1, 'FP16', output_shape=(1000))
#model = TftrtEngine(frozenmodel, 1, 'INT8', output_shape=(1000))
print('TF-TRT model ready to rumble!')
# Load and preprocess the image
input_image = PIL.Image.open(input_image_path)
input_image = np.asarray(input_image)
preprocessed = preprocess_input(input_image)
preprocessed = np.expand_dims(preprocessed, axis=0)
print('input tensor shape : ' + str(preprocessed.shape))
# This actually calls the inference
print("Warmup prediction")
output = model.infer(preprocessed)
print(decode_predictions(output))
time.sleep(1)
print("starting now (Jetson Nano)...")
s = time.time()
for i in range(0,250,1):
output = model.infer(preprocessed)
e = time.time()
print('elapsed : ' + str(e-s))
import img_processing as iproc
from keras.applications.mobilenetv2 import MobileNetV2, decode_predictions
if __name__=="__main__":
'''
学習済みモデルのロード
'''
model = MobileNetV2(weights='imagenet')
# model.summary()
'''
学習用画像のロード
'''
imgs, paths = iproc.load_imgs('./tmp/')
'''
学習済みモデルによる推論
'''
predict = model.predict(imgs)
predict = decode_predictions(predict, top=1)
'''
分類ラベルリスト
http://image-net.org/challenges/LSVRC/2012/browse-synsets
'''
for idx in range(len(predict)):
item, name = predict[idx], paths[idx]
print('Predicted: %s \t Name: %s'%(str(item), name))
Main
"""
if __name__ == '__main__':
# load the model
model = MobileNetV2(alpha=1.0)
print(model.summary())
exit()
# load an image from file
image = load_img('mug.jpg', target_size=(224, 224))
# convert the image pixels to a numpy array
image = img_to_array(image)
# reshape data for the model
image = image.reshape((1, image.shape[0], image.shape[1], image.shape[2]))
# prepare the image for the VGG model
image = preprocess_input(image)
# predict the probability across all output classes
for i in range(_iter):
raw_input('{} iteration, press any key to perform...'.format(str(i)))
yhat = model.predict(image)
# return if not iter
if not _iter: exit()
# convert the probabilities to class labels
label = decode_predictions(yhat)
# retrieve the most likely result, e.g. highest probability
label = label[0][0]
# print the classification
print('%s (%.2f%%)' % (label[1], label[2] * 100))
# done.
from keras.models import Model, load_model
import numpy as np
# Performs classification on the specified image
# Spits out preditions
model = MobileNetV2(input_shape=None,
alpha=1.0,
depth_multiplier=1,
include_top=True,
weights='imagenet',
input_tensor=None,
pooling=None,
classes=1000)
img_path = 'cheetah.jpg'
img = image.load_img(img_path, target_size=(224, 224))
x = image.img_to_array(img)
x = np.expand_dims(x, axis=0)
x = preprocess_input(x)
preds = model.predict(x)
model.summary()
# decode the results into a list of tuples (class, description, probability)
# (one such list for each sample in the batch)
print('Predicted:', decode_predictions(preds, top=3)[0])
image -= np.min(image)
image = np.minimum(image, 255)
cam = cv2.applyColorMap(np.uint8(255 * heatmap), cv2.COLORMAP_JET)
cam = np.float32(cam) + np.float32(image)
cam = 255 * cam / np.max(cam)
return np.uint8(cam), heatmap
# preprocessed_input = load_image(sys.argv[1])
preprocessed_input = load_image("./klib/image/cat_dog.png")
model = MobileNetV2(weights='imagenet')
predictions = model.predict(preprocessed_input)
top_1 = decode_predictions(predictions)[0][0]
print('Predicted class:')
print('%s (%s) with probability %.2f' % (top_1[1], top_1[0], top_1[2]))
predicted_class = np.argmax(predictions)
cam, heatmap = grad_cam(input_model=model,
image=preprocessed_input,
category_index=predicted_class,
layer_name="Conv_1",
nb_classes=1000)
fig, ax = plt.subplots(1, 1, figsize=(5, 5))
ax.imshow(cv2.cvtColor(cv2.imread('./klib/image/cat_dog.png'),
cv2.COLOR_BGR2RGB))
ax.imshow(heatmap, alpha=0.5)
ax.grid(linewidth=0.2)
input_image = np.asarray(input_image)
# In[11]:
# Use the MobileNet preprocessing function,
# and expand dimensions to create a batch of 1 image
preprocessed = preprocess_input(input_image)
preprocessed = np.expand_dims(preprocessed, axis=0)
print('input tensor shape : ' + str(preprocessed.shape))
# In[12]:
# Do 1 warmup prediction, this way we make sure everything is loaded as it should
print("warmup prediction")
prediction = net.predict(preprocessed)
print(decode_predictions(prediction, top=1)[0])
time.sleep(1)
# In[13]:
print("starting now...")
s = time.time()
for i in range(0, 250, 1):
prediction = net.predict(preprocessed)
e = time.time()
print('Time[ms] : ' + str(e - s))
print('FPS : ' + str(1.0 / ((e - s) / 250.0)))
# In[ ]: