识别细胞的细胞核是大多数分析的起点，因为人体的30万亿个细胞中，大多数都含有一个充满DNA的细胞核，DNA是编码每个细胞的遗传密码。识别细胞核使研究人员能够识别样本中的每个单独细胞，通过测量细胞对各种处理的反应，研究人员可以了解工作中的潜在生物过程。
加载必要的库和导入数据
import os
import sys
import random
import warnings
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from tqdm import tqdm
from itertools import chain
from skimage.io import imread, imshow, imread_collection, concatenate_images
from skimage.transform import resize
from skimage.morphology import label
from keras.models import Model, load_model
from keras.layers import Input
from keras.layers.core import Dropout, Lambda
from keras.layers.convolutional import Conv2D, Conv2DTranspose
from keras.layers.pooling import MaxPooling2D
from keras.layers.merge import concatenate
from keras.callbacks import EarlyStopping, ModelCheckpoint
from keras import backend as K
import tensorflow as tf
# 参数设置
IMG_WIDTH = 128
IMG_HEIGHT = 128
IMG_CHANNELS = 3
TRAIN_PATH = '../input/stage1_train/'
TEST_PATH = '../input/stage1_test/'
warnings.filterwarnings('ignore', category=UserWarning, module='skimage')
seed = 42
random.seed = seed
np.random.seed = seed
使用TensorFlow后端。
# Get train and test IDs
train_ids = next(os.walk(TRAIN_PATH))[1]
test_ids = next(os.walk(TEST_PATH))[1]
获取数据
让我们首先导入所有的图像和相关的掩码。我对训练图像和测试图像进行了降采样，以保持简单和易于管理，但我们需要保留测试图像的原始大小的记录，以对我们预测的掩码进行上采样，并在以后创建正确的游程编码。肯定有更好的方法来处理这个问题，但目前它工作得很好！
# 获取并调整训练图像和掩模的大小
X_train = np.zeros((len(train_ids), IMG_HEIGHT, IMG_WIDTH, IMG_CHANNELS), dtype=np.uint8)
Y_train = np.zeros((len(train_ids), IMG_HEIGHT, IMG_WIDTH, 1), dtype=np.bool)
print('Getting and resizing train images and masks ... ')
sys.stdout.flush()
for n, id_ in tqdm(enumerate(train_ids), total=len(train_ids)):
path = TRAIN_PATH + id_
img = imread(path + '/images/' + id_ + '.png')[:,:,:IMG_CHANNELS]
img = resize(img, (IMG_HEIGHT, IMG_WIDTH), mode='constant', preserve_range=True)
X_train[n] = img
mask = np.zeros((IMG_HEIGHT, IMG_WIDTH, 1), dtype=np.bool)
for mask_file in next(os.walk(path + '/masks/'))[2]:
mask_ = imread(path + '/masks/' + mask_file)
mask_ = np.expand_dims(resize(mask_, (IMG_HEIGHT, IMG_WIDTH), mode='constant',
preserve_range=True), axis=-1)
mask = np.maximum(mask, mask_)
Y_train[n] = mask
# 获取测试图像并调整其大小
X_test = np.zeros((len(test_ids), IMG_HEIGHT, IMG_WIDTH, IMG_CHANNELS), dtype=np.uint8)
sizes_test = []
print('Getting and resizing test images ... ')
sys.stdout.flush()
for n, id_ in tqdm(enumerate(test_ids), total=len(test_ids)):
path = TEST_PATH + id_
img = imread(path + '/images/' + id_ + '.png')[:,:,:IMG_CHANNELS]
sizes_test.append([img.shape[0], img.shape[1]])
img = resize(img, (IMG_HEIGHT, IMG_WIDTH), mode='constant', preserve_range=True)
X_test[n] = img
print('Done!')
让我们通过随机绘制一些图像和它们相关的掩码来看看是否一切正常。
# Check if training data looks all right
ix = random.randint(0, len(train_ids))
imshow(X_train[ix])
plt.show()
imshow(np.squeeze(Y_train[ix]))
plt.show()
创建Keras指标
现在我们尝试定义Keras中不同交并比（IoU）阈值的平均精度均值。TensorFlow有一个平均IoU指标，但它对多个阈值的平均值没有任何原生支持，所以我尝试实现它。但是，我绝不确定这个实现是正确的！任何协助验证这将是最受欢迎的！
更新：这个实现肯定是不正确的，因为这里报告的结果与LB结果之间有很大的差异。无论你在什么时候训练，它似乎都会随着时间的推移而增加……
# 定义IoU度量
def mean_iou(y_true, y_pred):
prec = []
for t in np.arange(0.5, 1.0, 0.05):
y_pred_ = tf.to_int32(y_pred > t)
score, up_opt = tf.metrics.mean_iou(y_true, y_pred_, 2)
K.get_session().run(tf.local_variables_initializer())
with tf.control_dependencies([up_opt]):
score = tf.identity(score)
prec.append(score)
return K.mean(K.stack(prec), axis=0)
构建并训练我们的神经网络
接下来，我们构建我们的U-Net模型，大致基于U-Net：用于生物医学图像分割的卷积网络
# 构建 U-Net model
inputs = Input((IMG_HEIGHT, IMG_WIDTH, IMG_CHANNELS))
s = Lambda(lambda x: x / 255) (inputs)
c1 = Conv2D(16, (3, 3), activation='elu', kernel_initializer='he_normal', padding='same') (s)
c1 = Dropout(0.1) (c1)
c1 = Conv2D(16, (3, 3), activation='elu', kernel_initializer='he_normal', padding='same') (c1)
p1 = MaxPooling2D((2, 2)) (c1)
c2 = Conv2D(32, (3, 3), activation='elu', kernel_initializer='he_normal', padding='same') (p1)
c2 = Dropout(0.1) (c2)
c2 = Conv2D(32, (3, 3), activation='elu', kernel_initializer='he_normal', padding='same') (c2)
p2 = MaxPooling2D((2, 2)) (c2)
c3 = Conv2D(64, (3, 3), activation='elu', kernel_initializer='he_normal', padding='same') (p2)
c3 = Dropout(0.2) (c3)
c3 = Conv2D(64, (3, 3), activation='elu', kernel_initializer='he_normal', padding='same') (c3)
p3 = MaxPooling2D((2, 2)) (c3)
c4 = Conv2D(128, (3, 3), activation='elu', kernel_initializer='he_normal', padding='same') (p3)
c4 = Dropout(0.2) (c4)
c4 = Conv2D(128, (3, 3), activation='elu', kernel_initializer='he_normal', padding='same') (c4)
p4 = MaxPooling2D(pool_size=(2, 2)) (c4)
c5 = Conv2D(256, (3, 3), activation='elu', kernel_initializer='he_normal', padding='same') (p4)
c5 = Dropout(0.3) (c5)
c5 = Conv2D(256, (3, 3), activation='elu', kernel_initializer='he_normal', padding='same') (c5)
u6 = Conv2DTranspose(128, (2, 2), strides=(2, 2), padding='same') (c5)
u6 = concatenate([u6, c4])
c6 = Conv2D(128, (3, 3), activation='elu', kernel_initializer='he_normal', padding='same') (u6)
c6 = Dropout(0.2) (c6)
c6 = Conv2D(128, (3, 3), activation='elu', kernel_initializer='he_normal', padding='same') (c6)
u7 = Conv2DTranspose(64, (2, 2), strides=(2, 2), padding='same') (c6)
u7 = concatenate([u7, c3])
c7 = Conv2D(64, (3, 3), activation='elu', kernel_initializer='he_normal', padding='same') (u7)
c7 = Dropout(0.2) (c7)
c7 = Conv2D(64, (3, 3), activation='elu', kernel_initializer='he_normal', padding='same') (c7)
u8 = Conv2DTranspose(32, (2, 2), strides=(2, 2), padding='same') (c7)
u8 = concatenate([u8, c2])
c8 = Conv2D(32, (3, 3), activation='elu', kernel_initializer='he_normal', padding='same') (u8)
c8 = Dropout(0.1) (c8)
c8 = Conv2D(32, (3, 3), activation='elu', kernel_initializer='he_normal', padding='same') (c8)
u9 = Conv2DTranspose(16, (2, 2), strides=(2, 2), padding='same') (c8)
u9 = concatenate([u9, c1], axis=3)
c9 = Conv2D(16, (3, 3), activation='elu', kernel_initializer='he_normal', padding='same') (u9)
c9 = Dropout(0.1) (c9)
c9 = Conv2D(16, (3, 3), activation='elu', kernel_initializer='he_normal', padding='same') (c9)
outputs = Conv2D(1, (1, 1), activation='sigmoid') (c9)
model = Model(inputs=[inputs], outputs=[outputs])
model.compile(optimizer='adam', loss='binary_crossentropy', metrics=[mean_iou])
model.summary()
更新：更改为ELU单位，添加dropout。
接下来在训练数据上拟合模型，验证划分为0.1。我们使用小批量，因为我们的数据非常少。我建议在训练模型时使用检查点和提前停止。我不会在这里这样做，以使事情更具有可重复性（尽管你的结果很可能会不同）。
# Fit model
earlystopper = EarlyStopping(patience=5, verbose=1)
checkpointer = ModelCheckpoint('model-dsbowl2018-1.h5', verbose=1, save_best_only=True)
results = model.fit(X_train, Y_train, validation_split=0.1, batch_size=16, epochs=50,
callbacks=[earlystopper, checkpointer])
作出预测
让我们对测试集、val集和训练集进行预测（作为完整性检查）。如果使用了提前停止和检查点机制，记得加载保存得最好的模型。
# 对train、val和test进行预测
model = load_model('model-dsbowl2018-1.h5', custom_objects={'mean_iou': mean_iou})
preds_train = model.predict(X_train[:int(X_train.shape[0]*0.9)], verbose=1)
preds_val = model.predict(X_train[int(X_train.shape[0]*0.9):], verbose=1)
preds_test = model.predict(X_test, verbose=1)
# 阈值预测
preds_train_t = (preds_train > 0.5).astype(np.uint8)
preds_val_t = (preds_val > 0.5).astype(np.uint8)
preds_test_t = (preds_test > 0.5).astype(np.uint8)
# 创建上采样测试掩码列表
preds_test_upsampled = []
for i in range(len(preds_test)):
preds_test_upsampled.append(resize(np.squeeze(preds_test[i]),
(sizes_test[i][0], sizes_test[i][1]),
mode='constant', preserve_range=True))
# 对一些随机训练样本进行完整性检查
ix = random.randint(0, len(preds_train_t))
imshow(X_train[ix])
plt.show()
imshow(np.squeeze(Y_train[ix]))
plt.show()
imshow(np.squeeze(preds_train_t[ix]))
plt.show()
该模型至少能够拟合训练数据！当然，即使在这里也有很大的改进空间，但这是一个不错的开始。验证数据呢？
# 对一些随机验证样本进行完整性检查
ix = random.randint(0, len(preds_val_t))
imshow(X_train[int(X_train.shape[0]*0.9):][ix])
plt.show()
imshow(np.squeeze(Y_train[int(Y_train.shape[0]*0.9):][ix]))
plt.show()
imshow(np.squeeze(preds_val_t[ix]))
plt.show()
编码并提交结果
def rle_encoding(x):
dots = np.where(x.T.flatten() == 1)[0]
run_lengths = []
prev = -2
for b in dots:
if (b>prev+1): run_lengths.extend((b + 1, 0))
run_lengths[-1] += 1
prev = b
return run_lengths
def prob_to_rles(x, cutoff=0.5):
lab_img = label(x > cutoff)
for i in range(1, lab_img.max() + 1):
yield rle_encoding(lab_img == i)
让我们遍历测试id并为skimage标识的每个单独掩码生成游程编码。
new_test_ids = []
rles = []
for n, id_ in enumerate(test_ids):
rle = list(prob_to_rles(preds_test_upsampled[n]))
rles.extend(rle)
new_test_ids.extend([id_] * len(rle))
# 创建提交DataFrame
sub = pd.DataFrame()
sub['ImageId'] = new_test_ids
sub['EncodedPixels'] = pd.Series(rles).apply(lambda x: ' '.join(str(y) for y in x))
sub.to_csv('sub-dsbowl2018-1.csv', index=False)
