深度学习 —— 个人学习笔记18（语义分割数据集、区域卷积神经网络）

声明

  本文章为个人学习使用，版面观感若有不适请谅解，文中知识仅代表个人观点，若出现错误，欢迎各位批评指正。

三十六、语义分割数据集

import os
import torch
import torchvision
import matplotlib.pyplot as plt

def show_images(imgs, num_rows, num_cols, titles=None, scale=1.5, suptitles=None):
    numpy = lambda x, *args, **kwargs: x.detach().numpy(*args, **kwargs)
    figsize = (num_cols * scale, num_rows * scale)
    _, axes = plt.subplots(num_rows, num_cols, figsize=figsize)
    axes = axes.flatten()
    for i, (ax, img) in enumerate(zip(axes, imgs)):
        try:
            img = numpy(img)
        except:
            pass
        ax.imshow(img)
        ax.axes.get_xaxis().set_visible(False)
        ax.axes.get_yaxis().set_visible(False)
        if titles:
            ax.set_title(titles[i])
    plt.rcParams['font.sans-serif'] = ['Microsoft YaHei']
    plt.suptitle(suptitles)
    plt.show()
    return axes

def read_voc_images(voc_dir, is_train=True):
    """读取所有VOC图像并标注"""
    txt_fname = os.path.join(voc_dir, 'ImageSets', 'Segmentation',
                             'train.txt' if is_train else 'val.txt')
    mode = torchvision.io.image.ImageReadMode.RGB
    with open(txt_fname, 'r') as f:
        images = f.read().split()
    features, labels = [], []
    for i, fname in enumerate(images):
        features.append(torchvision.io.read_image(os.path.join(
            voc_dir, 'JPEGImages', f'{fname}.jpg')))
        labels.append(torchvision.io.read_image(os.path.join(
            voc_dir, 'SegmentationClass' ,f'{fname}.png'), mode))
    return features, labels


voc_dir = 'E:\\VOCdevkit\\VOC2012'
train_features, train_labels = read_voc_images(voc_dir, True)

n = 5
imgs = train_features[0:n] + train_labels[0:n]
imgs = [img.permute(1,2,0) for img in imgs]
show_images(imgs, 2, n, suptitles='绘制前5个输入图像及其标签。\n标签图像中，白色和黑色分别表示边框和背景，其他颜色对应不同的类别')

VOC_COLORMAP = [[0, 0, 0], [128, 0, 0], [0, 128, 0], [128, 128, 0],
                [0, 0, 128], [128, 0, 128], [0, 128, 128], [128, 128, 128],
                [64, 0, 0], [192, 0, 0], [64, 128, 0], [192, 128, 0],
                [64, 0, 128], [192, 0, 128], [64, 128, 128], [192, 128, 128],
                [0, 64, 0], [128, 64, 0], [0, 192, 0], [128, 192, 0],
                [0, 64, 128]]

VOC_CLASSES = ['background', 'aeroplane', 'bicycle', 'bird', 'boat',
               'bottle', 'bus', 'car', 'cat', 'chair', 'cow',
               'diningtable', 'dog', 'horse', 'motorbike', 'person',
               'potted plant', 'sheep', 'sofa', 'train', 'tv/monitor']

def voc_colormap2label():
    """构建从RGB到VOC类别索引的映射"""
    colormap2label = torch.zeros(256 ** 3, dtype=torch.long)
    for i, colormap in enumerate(VOC_COLORMAP):
        colormap2label[
            (colormap[0] * 256 + colormap[1]) * 256 + colormap[2]] = i
    return colormap2label

def voc_label_indices(colormap, colormap2label):
    """将VOC标签中的RGB值映射到它们的类别索引"""
    colormap = colormap.permute(1, 2, 0).numpy().astype('int32')
    idx = ((colormap[:, :, 0] * 256 + colormap[:, :, 1]) * 256
           + colormap[:, :, 2])
    return colormap2label[idx]


y = voc_label_indices(train_labels[0], voc_colormap2label())
print("在第一张样本图像中，飞机头部区域的类别索引为 1，而背景索引为 0\n",
      y[105:115, 130:140], VOC_CLASSES[1])

def voc_rand_crop(feature, label, height, width):
    """随机裁剪特征和标签图像"""
    rect = torchvision.transforms.RandomCrop.get_params(
        feature, (height, width))
    feature = torchvision.transforms.functional.crop(feature, *rect)
    label = torchvision.transforms.functional.crop(label, *rect)
    return feature, label


imgs = []
for _ in range(n):
    imgs += voc_rand_crop(train_features[0], train_labels[0], 200, 300)


imgs = [img.permute(1, 2, 0) for img in imgs]
show_images(imgs[::2] + imgs[1::2], 2, n, suptitles='使用图像增广中随机裁剪后的效果')

class VOCSegDataset(torch.utils.data.Dataset):
    """一个用于加载VOC数据集的自定义数据集"""

    def __init__(self, is_train, crop_size, voc_dir):
        self.transform = torchvision.transforms.Normalize(
            mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
        self.crop_size = crop_size
        features, labels = read_voc_images(voc_dir, is_train=is_train)
        self.features = [self.normalize_image(feature)
                         for feature in self.filter(features)]
        self.labels = self.filter(labels)
        self.colormap2label = voc_colormap2label()
        if is_train:
            print('train : read ' + str(len(self.features)) + ' examples')
        else:
            print('test : read ' + str(len(self.features)) + ' examples')

    def normalize_image(self, img):
        return self.transform(img.float() / 255)

    def filter(self, imgs):
        return [img for img in imgs if (
            img.shape[1] >= self.crop_size[0] and
            img.shape[2] >= self.crop_size[1])]

    def __getitem__(self, idx):
        feature, label = voc_rand_crop(self.features[idx], self.labels[idx],
                                       *self.crop_size)
        return (feature, voc_label_indices(label, self.colormap2label))

    def __len__(self):
        return len(self.features)


crop_size = (320, 480)
voc_train = VOCSegDataset(True, crop_size, voc_dir)
voc_test = VOCSegDataset(False, crop_size, voc_dir)

batch_size = 64
train_iter = torch.utils.data.DataLoader(voc_train, batch_size, shuffle=True, drop_last=True)

for X, Y in train_iter:
    print(f'第一个小批量的 X 形状 : {X.shape}')
    print(f'第一个小批量的 Y 形状 : {Y.shape}')
    break

def load_data_voc(batch_size, crop_size):
    """加载VOC语义分割数据集"""
    voc_dir = 'E:\\VOCdevkit\\VOC2012'
    train_iter = torch.utils.data.DataLoader(
        VOCSegDataset(True, crop_size, voc_dir), batch_size,
        shuffle=True, drop_last=True)
    test_iter = torch.utils.data.DataLoader(
        VOCSegDataset(False, crop_size, voc_dir),
        batch_size, drop_last=True)
    return train_iter, test_iter


train_iter, test_iter = load_data_voc(batch_size, crop_size)
for X, Y in train_iter:
    print(f'第一个小批量的 X 形状 : {X.shape}')
    print(f'第一个小批量的 Y 形状 : {Y.shape}')
    break

三十七、区域卷积神经网络（R - CNN）

import torch
import torchvision
import matplotlib.pyplot as plt

def bbox_to_rect(bbox, color):
    return plt.Rectangle(
        xy=(bbox[0], bbox[1]), width=bbox[2]-bbox[0], height=bbox[3]-bbox[1],
        fill=False, edgecolor=color, linewidth=2)

def show_bboxes(axes, bboxes, labels=None, colors=None):
    """显示所有边界框"""
    def _make_list(obj, default_values=None):
        if obj is None:
            obj = default_values
        elif not isinstance(obj, (list, tuple)):
            obj = [obj]
        return obj

    labels = _make_list(labels)
    colors = _make_list(colors, ['b', 'g', 'r', 'm', 'c'])
    for i, bbox in enumerate(bboxes):
        color = colors[i % len(colors)]
        rect = bbox_to_rect(bbox.detach().numpy(), color)
        axes.add_patch(rect)
        if labels and len(labels) > i:
            text_color = 'k' if color == 'w' else 'w'
            axes.text(rect.xy[0], rect.xy[1], labels[i],
                      va='center', ha='center', fontsize=9, color=text_color,
                      bbox=dict(facecolor=color, lw=0))

def show_image(img, subplot, anchor_box=None, is_anchor=False, is_finally=False, titles=None):
    plt.subplot(subplot)
    plt.imshow(img, vmin=0, vmax=15)
    plt.colorbar()
    plt.rcParams['font.sans-serif'] = ['Microsoft YaHei']
    plt.title(titles)
    if is_anchor:
        show_bboxes(plt.imshow(img).axes, anchor_box, ['0', '1'])
    elif is_finally:
        plt.show()


X = torch.arange(16.).reshape(1, 1, 4, 4)
print(f'X : {X}')

rois = torch.Tensor([[0, 0, 0, 20, 20], [0, 0, 10, 30, 30]])
feature = torchvision.ops.roi_pool(X, rois, output_size=(2, 2), spatial_scale=0.1)
print(f'特定区域内特征 : {feature}')

plt.figure(figsize=(15.5, 6.5))
show_image(X.reshape(4, 4, 1), 131, titles='初始图像')
show_image(X.reshape(4, 4, 1), 132, anchor_box=rois[:, 1:] * 0.1, is_anchor=True, titles='框出特定区域')
show_image(feature.reshape(2, 4, 1), 133, titles='抽取特征', is_finally=True)

---

  文中部分知识参考：B 站 —— 跟李沐学AI；百度百科