目标检测(锚框)
目标检测需要给出物体的边缘框
一类目标检测算法是基于锚框(anchor box)的:
- 提出多个被称为锚框的区域(边缘框),是对物体位置的猜测
- 预测每个锚框里是否含有关注的物体
- 如果这个锚框中还有关注的物体,则预测从这个锚框到真实边缘框的偏移
1 锚框的生成
1.1 计算两个框之间的相似度(IoU交并比)
比如给定两个集合A和B,计算IoU过程如下:
对于图片来说,两个锚框的相似度就是重叠部分的像素值
1.2 赋予锚框标号
- 每个锚框是一个训练样本
- 将每个锚框要么标注为背景,要么关联上一个真实的边缘框,计算其偏移用于训练
- 我们可能会生成大量的锚框,这样会导致大量的负类样本
比如生成了9个锚框,识别的物体个数为4,标注过程如下
- 对每个锚框计算与每个物体边缘框的IoU,会生成一个矩阵,如上图所示
- 选出矩阵中最大值,比如是x23,它对应边缘框3,那么将该锚框用于预测边缘框3,然后删除它所在行和所在列的所有数据
- 对于剩下的数据,继续选择最大值,然后关联上边缘框,继续下去,直到将所有边缘框都关联上了锚框
值得注意的是:
- 每读取一张图片,都要生成一次锚框,比如9个锚框,然后对其进行标号,生成训练样本
- 赋予标号的算法有多种,上面只是其中常见的方法
1.3 使用非极大值抑制(NMS)输出
因为每个锚框会预测一个边缘框,其中有很多相似的预测,可以通过NMS来合并相似的预测
- 选中是非背景类的最大预测值(最大预测值是指对类的预测的softmax值)
- 去掉所有其他锚框和该锚框IoU值大于\theta的预测
- 重复上述过程,直到所有预测框要么被选中,要么被去掉
总结:
- 一类目标检测算法基于锚框来预测
- 首先生成大量锚框,并赋予标号,每个锚框作为一个样本进行训练
- 在预测时,使用NMS来取消冗余的预测
2 代码实现
2.1 锚框的生成和显示
锚框的生成(其中一种方法)
以每一个像素为中心,生成不同高宽度的锚框
- s 表示锚框大小,占图片的百分之多少
- r 锚框的高宽比
- 每次生成的时候会给出多个s和多个r,但是不会一一组合,因为会生成大量的锚框,通常一般做法是用一个最合适的s和每个r进行一次组合,然后将后续的s与r1进行组合
- 所以生成的锚框数量=s的个数+r的个数-1
%matplotlib inline
import torch
from d2l import torch as d2l
torch.set_printoptions(2) # 精简输出精度
#@save
# data 图片, size 锚框的大小,占图片的大小, ratio表示锚框的高宽比
def multibox_prior(data, sizes, ratios):
"""生成以每个像素为中心具有不同形状的锚框"""
in_height, in_width = data.shape[-2:]
device, num_sizes, num_ratios = data.device, len(sizes), len(ratios)
boxes_per_pixel = (num_sizes + num_ratios - 1)
size_tensor = torch.tensor(sizes, device=device)
ratio_tensor = torch.tensor(ratios, device=device)
# 为了将锚点移动到像素的中心,需要设置偏移量。
# 因为一个像素的高为1且宽为1,我们选择偏移我们的中心0.5
offset_h, offset_w = 0.5, 0.5
steps_h = 1.0 / in_height # 在y轴上缩放步长
steps_w = 1.0 / in_width # 在x轴上缩放步长
# 生成锚框的所有中心点
center_h = (torch.arange(in_height, device=device) + offset_h) * steps_h
center_w = (torch.arange(in_width, device=device) + offset_w) * steps_w
shift_y, shift_x = torch.meshgrid(center_h, center_w, indexing='ij')
shift_y, shift_x = shift_y.reshape(-1), shift_x.reshape(-1)
# 生成“boxes_per_pixel”个高和宽,
# 之后用于创建锚框的四角坐标(xmin,xmax,ymin,ymax)
w = torch.cat((size_tensor * torch.sqrt(ratio_tensor[0]),
sizes[0] * torch.sqrt(ratio_tensor[1:])))\
* in_height / in_width # 处理矩形输入
h = torch.cat((size_tensor / torch.sqrt(ratio_tensor[0]),
sizes[0] / torch.sqrt(ratio_tensor[1:])))
# 除以2来获得半高和半宽
anchor_manipulations = torch.stack((-w, -h, w, h)).T.repeat(
in_height * in_width, 1) / 2
# 每个中心点都将有“boxes_per_pixel”个锚框,
# 所以生成含所有锚框中心的网格,重复了“boxes_per_pixel”次
out_grid = torch.stack([shift_x, shift_y, shift_x, shift_y],
dim=1).repeat_interleave(boxes_per_pixel, dim=0)
output = out_grid + anchor_manipulations
return output.unsqueeze(0)
img = d2l.plt.imread('img/catdog.jpg')
h, w = img.shape[:2]
print(h, w)
X = torch.rand(size=(1, 3, h, w))
Y = multibox_prior(X, sizes=[0.75, 0.5, 0.25], ratios=[1, 2, 0.5])
Y.shape
输出:
561 728
torch.Size([1, 2042040, 4]) 生成了两百多万个锚框
访问以(250, 250)为中心的第一个锚框
boxes = Y.reshape(h, w, 5, 4)
boxes[250, 250, 0, :]
输出:
tensor([0.06, 0.07, 0.63, 0.82]) # 已经除以高宽了
显示以图像中一个像素为中心的所有锚框
#@save
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 = d2l.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))
d2l.set_figsize()
bbox_scale = torch.tensor((w, h, w, h))
fig = d2l.plt.imshow(img)
show_bboxes(fig.axes, boxes[250, 250, :, :] * bbox_scale,
['s=0.75, r=1', 's=0.5, r=1', 's=0.25, r=1', 's=0.75, r=2',
's=0.75, r=0.5'])
2.2 给锚框进行标号
计算两个框的交并比
#@save
def box_iou(boxes1, boxes2):
"""计算两个锚框或边界框列表中成对的交并比"""
box_area = lambda boxes: ((boxes[:, 2] - boxes[:, 0]) *
(boxes[:, 3] - boxes[:, 1]))
# boxes1,boxes2,areas1,areas2的形状:
# boxes1:(boxes1的数量,4),
# boxes2:(boxes2的数量,4),
# areas1:(boxes1的数量,),
# areas2:(boxes2的数量,)
areas1 = box_area(boxes1)
areas2 = box_area(boxes2)
# inter_upperlefts,inter_lowerrights,inters的形状:
# (boxes1的数量,boxes2的数量,2)
inter_upperlefts = torch.max(boxes1[:, None, :2], boxes2[:, :2])
inter_lowerrights = torch.min(boxes1[:, None, 2:], boxes2[:, 2:])
inters = (inter_lowerrights - inter_upperlefts).clamp(min=0)
# inter_areasandunion_areas的形状:(boxes1的数量,boxes2的数量)
inter_areas = inters[:, :, 0] * inters[:, :, 1]
union_areas = areas1[:, None] + areas2 - inter_areas
return inter_areas / union_areas
将最接近的真实边缘框分配给锚框
#@save
def assign_anchor_to_bbox(ground_truth, anchors, device, iou_threshold=0.5):
"""将最接近的真实边界框分配给锚框"""
num_anchors, num_gt_boxes = anchors.shape[0], ground_truth.shape[0]
# 位于第i行和第j列的元素x_ij是锚框i和真实边界框j的IoU
jaccard = box_iou(anchors, ground_truth)
# 对于每个锚框,分配的真实边界框的张量
anchors_bbox_map = torch.full((num_anchors,), -1, dtype=torch.long,
device=device)
# 根据阈值,决定是否分配真实边界框
max_ious, indices = torch.max(jaccard, dim=1)
anc_i = torch.nonzero(max_ious >= iou_threshold).reshape(-1)
box_j = indices[max_ious >= iou_threshold]
anchors_bbox_map[anc_i] = box_j
col_discard = torch.full((num_anchors,), -1)
row_discard = torch.full((num_gt_boxes,), -1)
for _ in range(num_gt_boxes):
max_idx = torch.argmax(jaccard)
box_idx = (max_idx % num_gt_boxes).long()
anc_idx = (max_idx / num_gt_boxes).long()
anchors_bbox_map[anc_idx] = box_idx
jaccard[:, box_idx] = col_discard
jaccard[anc_idx, :] = row_discard
return anchors_bbox_map
标记类和偏移
#@save
# 让机器学习算法更好的预测,计算偏移
def offset_boxes(anchors, assigned_bb, eps=1e-6):
"""对锚框偏移量的转换"""
c_anc = d2l.box_corner_to_center(anchors)
c_assigned_bb = d2l.box_corner_to_center(assigned_bb)
offset_xy = 10 * (c_assigned_bb[:, :2] - c_anc[:, :2]) / c_anc[:, 2:]
offset_wh = 5 * torch.log(eps + c_assigned_bb[:, 2:] / c_anc[:, 2:])
offset = torch.cat([offset_xy, offset_wh], axis=1)
return offset
# 这个用来返回类别和偏移
#@save
def multibox_target(anchors, labels):
"""使用真实边界框标记锚框"""
batch_size, anchors = labels.shape[0], anchors.squeeze(0)
batch_offset, batch_mask, batch_class_labels = [], [], []
device, num_anchors = anchors.device, anchors.shape[0]
for i in range(batch_size):
label = labels[i, :, :]
anchors_bbox_map = assign_anchor_to_bbox(
label[:, 1:], anchors, device)
bbox_mask = ((anchors_bbox_map >= 0).float().unsqueeze(-1)).repeat(
1, 4)
# 将类标签和分配的边界框坐标初始化为零
class_labels = torch.zeros(num_anchors, dtype=torch.long,
device=device)
assigned_bb = torch.zeros((num_anchors, 4), dtype=torch.float32,
device=device)
# 使用真实边界框来标记锚框的类别。
# 如果一个锚框没有被分配,标记其为背景(值为零)
indices_true = torch.nonzero(anchors_bbox_map >= 0)
bb_idx = anchors_bbox_map[indices_true]
class_labels[indices_true] = label[bb_idx, 0].long() + 1
assigned_bb[indices_true] = label[bb_idx, 1:]
# 偏移量转换
offset = offset_boxes(anchors, assigned_bb) * bbox_mask
batch_offset.append(offset.reshape(-1))
batch_mask.append(bbox_mask.reshape(-1))
batch_class_labels.append(class_labels)
bbox_offset = torch.stack(batch_offset)
bbox_mask = torch.stack(batch_mask)
class_labels = torch.stack(batch_class_labels)
# bbox_offest 表示每一个anchor到真实框的offset
# bbox_mask=0 表示这个锚框是背景框,不用进行预测, 等于1表示对应一个真实的边框
# class_label表示锚框对应的类别标号
return (bbox_offset, bbox_mask, class_labels)
给个样例显示一下:
ground_truth = torch.tensor([[0, 0.1, 0.08, 0.52, 0.92],
[1, 0.55, 0.2, 0.9, 0.88]])
anchors = torch.tensor([[0, 0.1, 0.2, 0.3], [0.15, 0.2, 0.4, 0.4],
[0.63, 0.05, 0.88, 0.98], [0.66, 0.45, 0.8, 0.8],
[0.57, 0.3, 0.92, 0.9]])
fig = d2l.plt.imshow(img)
show_bboxes(fig.axes, ground_truth[:, 1:] * bbox_scale, ['dog', 'cat'], 'k')
show_bboxes(fig.axes, anchors * bbox_scale, ['0', '1', '2', '3', '4']);
运行上述函数
labels = multibox_target(anchors.unsqueeze(dim=0),
ground_truth.unsqueeze(dim=0))
labels[2] # 输出类别
输出:tensor([[0, 1, 2, 0, 2]]) #0表示背景,1表示狗,2表示猫
2.3 利用NMS进行预测
根据带有预测偏移的锚框来预测边缘框
#@save
def offset_inverse(anchors, offset_preds):
"""根据带有预测偏移量的锚框来预测边界框"""
anc = d2l.box_corner_to_center(anchors)
pred_bbox_xy = (offset_preds[:, :2] * anc[:, 2:] / 10) + anc[:, :2]
pred_bbox_wh = torch.exp(offset_preds[:, 2:] / 5) * anc[:, 2:]
pred_bbox = torch.cat((pred_bbox_xy, pred_bbox_wh), axis=1)
predicted_bbox = d2l.box_center_to_corner(pred_bbox)
return predicted_bbox
NMS算法
#@save
def nms(boxes, scores, iou_threshold):
"""对预测边界框的置信度进行排序"""
B = torch.argsort(scores, dim=-1, descending=True)
keep = [] # 保留预测边界框的指标
while B.numel() > 0:
i = B[0]
keep.append(i)
if B.numel() == 1: break
iou = box_iou(boxes[i, :].reshape(-1, 4),
boxes[B[1:], :].reshape(-1, 4)).reshape(-1)
inds = torch.nonzero(iou <= iou_threshold).reshape(-1)
B = B[inds + 1]
return torch.tensor(keep, device=boxes.device)
利用NMS来预测边缘框
#@save
def multibox_detection(cls_probs, offset_preds, anchors, nms_threshold=0.5,
pos_threshold=0.009999999):
"""使用非极大值抑制来预测边界框"""
device, batch_size = cls_probs.device, cls_probs.shape[0]
anchors = anchors.squeeze(0)
num_classes, num_anchors = cls_probs.shape[1], cls_probs.shape[2]
out = []
for i in range(batch_size):
cls_prob, offset_pred = cls_probs[i], offset_preds[i].reshape(-1, 4)
conf, class_id = torch.max(cls_prob[1:], 0)
predicted_bb = offset_inverse(anchors, offset_pred)
keep = nms(predicted_bb, conf, nms_threshold)
# 找到所有的non_keep索引,并将类设置为背景
all_idx = torch.arange(num_anchors, dtype=torch.long, device=device)
combined = torch.cat((keep, all_idx))
uniques, counts = combined.unique(return_counts=True)
non_keep = uniques[counts == 1]
all_id_sorted = torch.cat((keep, non_keep))
class_id[non_keep] = -1
class_id = class_id[all_id_sorted]
conf, predicted_bb = conf[all_id_sorted], predicted_bb[all_id_sorted]
# pos_threshold是一个用于非背景预测的阈值
below_min_idx = (conf < pos_threshold)
class_id[below_min_idx] = -1
conf[below_min_idx] = 1 - conf[below_min_idx]
pred_info = torch.cat((class_id.unsqueeze(1),
conf.unsqueeze(1),
predicted_bb), dim=1)
out.append(pred_info)
return torch.stack(out)
演示一下:
anchors = torch.tensor([[0.1, 0.08, 0.52, 0.92], [0.08, 0.2, 0.56, 0.95],
[0.15, 0.3, 0.62, 0.91], [0.55, 0.2, 0.9, 0.88]])
offset_preds = torch.tensor([0] * anchors.numel())
cls_probs = torch.tensor([[0] * 4, # 背景的预测概率
[0.9, 0.8, 0.7, 0.1], # 狗的预测概率
[0.1, 0.2, 0.3, 0.9]]) # 猫的预测概率
fig = d2l.plt.imshow(img)
show_bboxes(fig.axes, anchors * bbox_scale,
['dog=0.9', 'dog=0.8', 'dog=0.7', 'cat=0.9'])
output = multibox_detection(cls_probs.unsqueeze(dim=0),
offset_preds.unsqueeze(dim=0),
anchors.unsqueeze(dim=0),
nms_threshold=0.5)
fig = d2l.plt.imshow(img)
for i in output[0].detach().numpy():
if i[0] == -1:
continue
label = ('dog=', 'cat=')[int(i[0])] + str(i[1])
show_bboxes(fig.axes, [torch.tensor(i[2:]) * bbox_scale], label)