def __getitem__(self, idx):
self.get_lane_image(idx)
self.img, self.label_img = crop_resize_data(self.img, self.label_img)
self.img = CLAHE_nomalization(self.img)
self.img = contrast(self.img)
self.img, self.label_img = random_filp(self.img, self.label_img)
self.label_img = self.label_img * (self.label_img <
9) * (self.label_img >= 0)
self.get_ins_img()
# cv2.imshow('img', self.img)
# cv2.imshow('label', self.label_img)
# cv2.imshow('ins', self.ins_img[2]*100)
# cv2.waitKey(0)
# cv2.destroyAllWindows()
if self.flags['train']:
# self.img = np.array(np.transpose(self.img, (2,0,1)), dtype=np.float32)
self.img = self.transforms(self.img)
self.label_img = np.array(self.label_img, dtype=np.float32)
self.ins_img = np.array(self.ins_img, dtype=np.float32)
return torch.Tensor(self.img), torch.LongTensor(
self.label_img), torch.Tensor(self.ins_img)
else:
self.img = np.array(np.transpose(self.img, (2, 0, 1)),
dtype=np.float32)
return torch.Tensor(self.img)
def train_image_gen(train_list, batch_size=4, image_size=[1024, 384], crop_offset=690):
# Arrange all indexes
all_batches_index = np.arange(0, len(train_list))
out_images = []
out_masks = []
image_dir = np.array(train_list['image'])
label_dir = np.array(train_list['label'])
while True:
# Random shuffle indexes every epoch
np.random.shuffle(all_batches_index)
for index in all_batches_index:
if os.path.exists(image_dir[index]):
ori_image = cv2.imread(image_dir[index])
ori_mask = cv2.imread(label_dir[index], cv2.IMREAD_GRAYSCALE)
# Crop the top part of the image
# Resize to train size
train_img, train_mask = crop_resize_data(ori_image, ori_mask, image_size, crop_offset)
# Encode
train_mask = encode_labels(train_mask)
# verify_labels(train_mask)
out_images.append(train_img)
out_masks.append(train_mask)
if len(out_images) >= batch_size:
out_images = np.array(out_images)
out_masks = np.array(out_masks)
out_images = out_images[:, :, :, ::-1].transpose(0, 3, 1, 2).astype(np.float32) / (255.0 / 2) - 1
out_masks = out_masks.astype(np.int64)
yield out_images, out_masks
out_images, out_masks = [], []
else:
print(image_dir, 'does not exist.')
def img_transform(img):
img = crop_resize_data(img)
img = np.transpose(img, (2, 0, 1))
img = img[np.newaxis, ...].astype(np.float32)
img = torch.from_numpy(img.copy())
if torch.cuda.is_available():
img = img.cuda(device=device_id)
return img
def img_transform(img):
img = crop_resize_data(img)
# 轴
img = np.transpose(img, (2, 0, 1))
# batch的封装，添加了一个新的轴，在前面扩充一维，变成了 1*C*H*W,...不管图片后面哪一维，全都要
img = img[np.newaxis, ...].astype(np.float32)
img = torch.from_numpy(img.copy())
if torch.cuda.is_available():
img = img.cuda(device=device_id)
return img
def main():
IMG_SIZE =[1536, 512]
SUBMISSION_SIZE = [3384, 1710]
save_test_logits = False
num_classes = 8
batch_size = 4
log_iters = 100
network = 'unet_simple'
# Define paths for each model
if network == 'deeplabv3p':
model_path = "./model_weights/paddle_deeplabv3p_8_end_060223"
npy_dir = '/npy_save/deeplabv3p/'
elif network == 'unet_base':
model_path = "./model_weights/paddle_unet_base_10_end_059909"
npy_dir = '/npy_save/unet_base/'
elif network == 'unet_simple':
model_path = "./model_weights/paddle_unet_simple_12_end_060577"
npy_dir = '/npy_save/unet_simple/'
program_choice = 2 # 1 - Validtion; 2 - Test
show_label = False
crop_offset = 690
data_dir = './data_list/val.csv'
test_dir = '/root/private/LaneDataSet/TestSet/Image_Data/ColorImage/'
sub_dir = './test_submission/'
# Get data list and split it into train and validation set.
val_list = pd.read_csv(data_dir)
iter_id = 0
total_loss = 0.0
total_miou = 0.0
prev_time = time.time()
# Validation
if program_choice == 1:
val_reader = val_image_gen(val_list, batch_size=batch_size, image_size=IMG_SIZE, crop_offset=crop_offset)
model = create_network(network=network)
model.load_weights(model_path)
print("loaded model from: %s" % model_path)
print('Start Validation!')
for iteration in range(int(len(val_list) / batch_size)):
val_data,val_label = next(val_reader)
results = model.evaluate(val_data,val_label)
if iter_id % log_iters == 0:
print('Finished Processing %d Images.' %(iter_id * batch_size))
iter_id += 1
total_loss += np.mean(results[0])
total_miou += np.mean(results[1])
# label to mask
if show_label == True:
label_image = val_label[0]
color_label_mask = decode_color_labels(label_image)
color_label_mask = np.transpose(color_label_mask, (1, 2, 0))
cv2.imshow('gt_label', cv2.resize(color_label_mask, (IMG_SIZE[0], IMG_SIZE[1])))
prediction = np.argmax(results[2][0], axis=0)
color_pred_mask = decode_color_labels(prediction)
color_pred_mask = np.transpose(color_pred_mask, (1, 2, 0))
cv2.imshow('pred_label', cv2.resize(color_pred_mask, (IMG_SIZE[0], IMG_SIZE[1])))
cv2.waitKey(0)
end_time = time.time()
print("validation loss: %.3f, mean iou: %.3f, time cost: %.3f s"
% (total_loss / iter_id, total_miou / iter_id, end_time - prev_time))
# Test
elif program_choice == 2:
model = create_network(network=network)
model.load_weights(model_path)
print("loaded model from: %s" % model_path)
print('Start Making Submissions!')
test_list = os.listdir(test_dir)
for test_name in test_list:
test_ori_image = cv2.imread(os.path.join(test_dir, test_name))
test_image = crop_resize_data(test_ori_image, label=None, image_size=IMG_SIZE, offset=crop_offset)
out_image = np.expand_dims(np.array(test_image), axis=0)
out_image = out_image[:, :, :, ::-1].transpose(0, 3, 1, 2).astype(np.float32) / (255.0 / 2) - 1
results_1 = model.evaluate(out_image,None)
if iter_id % 20 == 0:
print('Finished Processing %d Images.' %(iter_id))
iter_id += 1
prediction = np.argmax(results_1[0][0], axis=0)
# Save npy files
if save_test_logits == True:
np.save(npy_dir + test_name.replace('.jpg', '.npy'), results_1[0][0])
# Save Submission PNG
submission_mask = expand_resize_data(prediction, SUBMISSION_SIZE, crop_offset)
cv2.imwrite(os.path.join(sub_dir, test_name.replace('.jpg', '.png')), submission_mask)
# Show Label
if show_label == True:
cv2.imshow('test_image', cv2.resize(test_ori_image,(IMG_SIZE[0], IMG_SIZE[1])))
cv2.imshow('pred_label', cv2.resize(submission_mask,(IMG_SIZE[0], IMG_SIZE[1])))
cv2.waitKey(0)
sys.stdout.flush()
import os
import cv2
import numpy as np
import pandas as pd
import imgaug as ia
from imgaug import augmenters as iaa
import albumentations as albu
import matplotlib.pyplot as plt
from utils.image_process import crop_resize_data
data = pd.read_csv(os.path.join(os.getcwd(), "data_list", "train.csv"), header=None,
names=["image", "label"])
images = data["image"].values[1:]
labels = data["label"].values[1:]
ori_image = cv2.imread(images[0])
ori_mask = cv2.imread(labels[0], cv2.IMREAD_GRAYSCALE)
ori_image, ori_mask = crop_resize_data(ori_image, ori_mask)
seq = iaa.Sequential([iaa.CropAndPad(
percent=(-0.05, 0.1))])
seg_to = seq.to_deterministic()
for i in range(8):
seg_to = seq.to_deterministic()
ori_image = seg_to.augment_image(ori_image) # 将方法应用在原图像上
ori_mask = seg_to.augment_image(ori_mask)
print(ori_mask.dtype)
print(ori_mask.shape)
plt.imshow(ori_mask)
plt.show()
def main():
IMG_SIZE =[1536, 512]
SUBMISSION_SIZE = [3384, 1710]
save_test_logits = False
num_classes = 8
batch_size = 4
log_iters = 100
network = 'unet_simple'
# Define paths for each model
if network == 'deeplabv3p':
model_path = "./model_weights/paddle_deeplabv3p_8_end_060223"
npy_dir = '/npy_save/deeplabv3p/'
elif network == 'unet_base':
model_path = "./model_weights/paddle_unet_base_10_end_059909"
npy_dir = '/npy_save/unet_base/'
elif network == 'unet_simple':
model_path = "./model_weights/paddle_unet_simple_12_end_060577"
npy_dir = '/npy_save/unet_simple/'
program_choice = 2 # 1 - Validtion; 2 - Test
show_label = False
crop_offset = 690
data_dir = './data_list/val.csv'
test_dir = '../PaddlePaddle/TestSet_Final/ColorImage/'
sub_dir = './test_submission/'
# Get data list and split it into train and validation set.
val_list = pd.read_csv(data_dir)
#Initialization
images = fluid.layers.data(name='image', shape=[3, IMG_SIZE[1], IMG_SIZE[0]], dtype='float32')
labels = fluid.layers.data(name='label', shape=[1, IMG_SIZE[1], IMG_SIZE[0]], dtype='float32')
iter_id = 0
total_loss = 0.0
total_miou = 0.0
prev_time = time.time()
# Validation
if program_choice == 1:
val_reader = val_image_gen(val_list, batch_size=batch_size, image_size=IMG_SIZE, crop_offset=crop_offset)
reduced_loss, miou, pred = create_network(images, labels, num_classes, network=network, image_size=(IMG_SIZE[1], IMG_SIZE[0]), for_test=False)
place = fluid.CUDAPlace(0)
exe = fluid.Executor(place)
exe.run(fluid.default_startup_program())
fluid.io.load_params(exe, model_path)
print("loaded model from: %s" % model_path)
# Parallel Executor to use multi-GPUs
exec_strategy = fluid.ExecutionStrategy()
exec_strategy.allow_op_delay = True
build_strategy = fluid.BuildStrategy()
build_strategy.reduce_strategy = fluid.BuildStrategy.ReduceStrategy.Reduce
train_exe = fluid.ParallelExecutor(use_cuda=True, build_strategy=build_strategy, exec_strategy=exec_strategy)
print('Start Validation!')
for iteration in range(int(len(val_list) / batch_size)):
val_data = next(val_reader)
results = train_exe.run(
feed=get_feeder_data(val_data, place),
fetch_list=[reduced_loss.name, miou.name, pred.name])
if iter_id % log_iters == 0:
print('Finished Processing %d Images.' %(iter_id * batch_size))
iter_id += 1
total_loss += np.mean(results[0])
total_miou += np.mean(results[1])
# label to mask
if show_label == True:
label_image = val_data[1][0]
color_label_mask = decode_color_labels(label_image)
color_label_mask = np.transpose(color_label_mask, (1, 2, 0))
cv2.imshow('gt_label', cv2.resize(color_label_mask, (IMG_SIZE[0], IMG_SIZE[1])))
prediction = np.argmax(results[2][0], axis=0)
color_pred_mask = decode_color_labels(prediction)
color_pred_mask = np.transpose(color_pred_mask, (1, 2, 0))
cv2.imshow('pred_label', cv2.resize(color_pred_mask, (IMG_SIZE[0], IMG_SIZE[1])))
cv2.waitKey(0)
end_time = time.time()
print("validation loss: %.3f, mean iou: %.3f, time cost: %.3f s"
% (total_loss / iter_id, total_miou / iter_id, end_time - prev_time))
# Test
elif program_choice == 2:
predictions = create_network(images, labels, num_classes, network=network, image_size=(IMG_SIZE[1], IMG_SIZE[0]), for_test=True)
place = fluid.CUDAPlace(0)
# place = fluid.CPUPlace()
exe = fluid.Executor(place)
exe.run(fluid.default_startup_program())
fluid.io.load_params(exe, model_path)
print("loaded model from: %s" % model_path)
print('Start Making Submissions!')
test_list = os.listdir(test_dir)
for test_name in test_list:
test_ori_image = cv2.imread(os.path.join(test_dir, test_name))
test_image = crop_resize_data(test_ori_image, label=None, image_size=IMG_SIZE, offset=crop_offset)
out_image = np.expand_dims(np.array(test_image), axis=0)
out_image = out_image[:, :, :, ::-1].transpose(0, 3, 1, 2).astype(np.float32) / (255.0 / 2) - 1
feed_dict = {}
feed_dict["image"] = out_image
results_1 = exe.run(
feed=feed_dict,
fetch_list=[predictions])
if iter_id % 20 == 0:
print('Finished Processing %d Images.' %(iter_id))
iter_id += 1
prediction = np.argmax(results_1[0][0], axis=0)
# Save npy files
if save_test_logits == True:
np.save(npy_dir + test_name.replace('.jpg', '.npy'), results_1[0][0])
# Save Submission PNG
submission_mask = expand_resize_data(prediction, SUBMISSION_SIZE, crop_offset)
cv2.imwrite(os.path.join(sub_dir, test_name.replace('.jpg', '.png')), submission_mask)
# Show Label
if show_label == True:
cv2.imshow('test_image', cv2.resize(test_ori_image,(IMG_SIZE[0], IMG_SIZE[1])))
cv2.imshow('pred_label', cv2.resize(submission_mask,(IMG_SIZE[0], IMG_SIZE[1])))
cv2.waitKey(0)
sys.stdout.flush()
