用于语义分割的加权像素明智分类交叉熵

Question

我最近开始学习语义分割。我正在尝试为此训练一个 UNet。我的输入是 RGB 128x128x3 图像。我的面具由 4 类 0、1、2、3 组成，并且是单热编码的，尺寸为 128x128x4。

def weighted_cce(y_true, y_pred):
        weights = []
        t_inf = tf.convert_to_tensor(1e9, dtype = 'float32')
        t_zero = tf.convert_to_tensor(0, dtype = 'int64')
        for i in range(0, 4):
            l = tf.argmax(y_true, axis = -1) == i
            n = tf.cast(tf.math.count_nonzero(l), 'float32') + K.epsilon()
            weights.append(n)

        weights = [batch_size/j for j in weights]

        y_pred /= K.sum(y_pred, axis=-1, keepdims=True)
        # clip to prevent NaN's and Inf's
        y_pred = K.clip(y_pred, K.epsilon(), 1 - K.epsilon())
        # calc
        loss = y_true * K.log(y_pred) * weights
        loss = -K.sum(loss, -1)
        return loss

这是我正在使用的损失函数，但它将每个像素分类为 2。我做错了什么？

Answer 1

你应该有基于你整个数据的权重（除非你的批量大小相当大所以你有稳定的权重）。

如果某些 class 未被充分代表，在小批量的情况下，它将具有接近无穷大的权重。

如果你的目标数据是numpy数组：

shp = y_train.shape
totalPixels = shp[0] * shp[1] * shp[2]

weights = np.sum(y_train, axis=(0, 1, 2)) #final shape (4,)
weights = totalPixels/weights

如果您的数据在 Sequence 生成器中：

totalPixels = 0
counts = np.zeros((4,))

for i in range(len(generator)):
    x, y = generator[i]

    shp = y.shape
    totalPixels += shp[0] * shp[1] * shp[2]

    counts = counts + np.sum(y, axis=(0,1,2))

weights = totalPixels / counts

如果你的数据在 yield 生成器中（你必须知道你在一个 epoch 中有多少批次）：

for i in range(batches_per_epoch):
    x, y = next(generator)
    #the rest is equal to the Sequence example above

尝试 1

我不知道更新版本的 Keras 是否能够处理这个问题，但您可以先尝试最简单的方法：只需使用 class_weight 调用 fit 或 fit_generator参数：

model.fit(...., class_weight = {0: weights[0], 1: weights[1], 2: weights[2], 3: weights[3]})

尝试 2

做一个更健康的损失函数：

weights = weights.reshape((1,1,1,4))
kWeights = K.constant(weights)

def weighted_cce(y_true, y_pred):
    yWeights = kWeights * y_pred         #shape (batch, 128, 128, 4)
    yWeights = K.sum(yWeights, axis=-1)  #shape (batch, 128, 128)  

    loss = K.categorical_crossentropy(y_true, y_pred) #shape (batch, 128, 128)
    wLoss = yWeights * loss

    return K.sum(wLoss, axis=(1,2))

用于语义分割的加权像素明智分类交叉熵

Weighted Pixel Wise Categorical Cross Entropy for Semantic Segmentation

image-segmentation

deep-learning

keras

tensorflow

semantic-segmentation

尝试 1

尝试 2