本文运行在GitHub上下载的YOLOv4代码,使用pytorch框架,在文章尾部贴出运行结果。
YOLOV4是YOLOV3的改进,相对于YOLOV3来说,YOLOV4做到如下改进:
注:我参考的博客是结合YOLO V3 和 V4的差别进行解析的,因此在此粘贴一个讲解YOLO V3的博客,供以参考:https://blog.csdn.net/weixin_44791964/article/details/105310627
注意,虽然是pytorch框架,但是本文的描述,都将通道数放在最后一个维度。
当输入是416 x 416 时,结构图如下:
当输入是608 x 608 x 3 时, 特征结构如下图所示:
关于主干特征提取网络的改进主要如下:
首先对图中左半部分的Resblock_body进行介绍,其为 一次 下采样 和 多次 残差结构的堆叠 构成。而Dark_net 53即由 resblock_body 模块组合而成。
在YOLO V4中,进行如下修改:
#---------------------------------------------------#
# CSPdarknet的结构块
# 存在一个大残差边
# 这个大残差边绕过了很多的残差结构
#---------------------------------------------------#
class Resblock_body(nn.Module):
def __init__(self, in_channels, out_channels, num_blocks, first):
super(Resblock_body, self).__init__()
self.downsample_conv = BasicConv(in_channels, out_channels, 3, stride=2)
if first:
self.split_conv0 = BasicConv(out_channels, out_channels, 1)
self.split_conv1 = BasicConv(out_channels, out_channels, 1)
self.blocks_conv = nn.Sequential(
Resblock(channels=out_channels, hidden_channels=out_channels//2),
BasicConv(out_channels, out_channels, 1)
)
self.concat_conv = BasicConv(out_channels*2, out_channels, 1)
else:
self.split_conv0 = BasicConv(out_channels, out_channels//2, 1)
self.split_conv1 = BasicConv(out_channels, out_channels//2, 1)
self.blocks_conv = nn.Sequential(
*[Resblock(out_channels//2) for _ in range(num_blocks)],
BasicConv(out_channels//2, out_channels//2, 1)
)
self.concat_conv = BasicConv(out_channels, out_channels, 1)
def forward(self, x):
x = self.downsample_conv(x)
x0 = self.split_conv0(x)
x1 = self.split_conv1(x)
x1 = self.blocks_conv(x1)
x = torch.cat([x1, x0], dim=1)
x = self.concat_conv(x)
return x
import torch
import torch.nn.functional as F
import torch.nn as nn
import math
from collections import OrderedDict
#-------------------------------------------------#
# MISH激活函数
#-------------------------------------------------#
class Mish(nn.Module):
def __init__(self):
super(Mish, self).__init__()
def forward(self, x):
return x * torch.tanh(F.softplus(x))
#-------------------------------------------------#
# 卷积块
# CONV+BATCHNORM+MISH
#-------------------------------------------------#
class BasicConv(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride=1):
super(BasicConv, self).__init__()
self.conv = nn.Conv2d(in_channels, out_channels, kernel_size, stride, kernel_size//2, bias=False)
self.bn = nn.BatchNorm2d(out_channels)
self.activation = Mish()
def forward(self, x):
x = self.conv(x)
x = self.bn(x)
x = self.activation(x)
return x
#---------------------------------------------------#
# CSPdarknet的结构块的组成部分
# 内部堆叠的残差块
#---------------------------------------------------#
class Resblock(nn.Module):
def __init__(self, channels, hidden_channels=None, residual_activation=nn.Identity()):
super(Resblock, self).__init__()
if hidden_channels is None:
hidden_channels = channels
self.block = nn.Sequential(
BasicConv(channels, hidden_channels, 1),
BasicConv(hidden_channels, channels, 3)
)
def forward(self, x):
return x+self.block(x)
#---------------------------------------------------#
# CSPdarknet的结构块
# 存在一个大残差边
# 这个大残差边绕过了很多的残差结构
#---------------------------------------------------#
class Resblock_body(nn.Module):
def __init__(self, in_channels, out_channels, num_blocks, first):
super(Resblock_body, self).__init__()
self.downsample_conv = BasicConv(in_channels, out_channels, 3, stride=2)
if first:
self.split_conv0 = BasicConv(out_channels, out_channels, 1)
self.split_conv1 = BasicConv(out_channels, out_channels, 1)
self.blocks_conv = nn.Sequential(
Resblock(channels=out_channels, hidden_channels=out_channels//2),
BasicConv(out_channels, out_channels, 1)
)
self.concat_conv = BasicConv(out_channels*2, out_channels, 1)
else:
self.split_conv0 = BasicConv(out_channels, out_channels//2, 1)
self.split_conv1 = BasicConv(out_channels, out_channels//2, 1)
self.blocks_conv = nn.Sequential(
*[Resblock(out_channels//2) for _ in range(num_blocks)],
BasicConv(out_channels//2, out_channels//2, 1)
)
self.concat_conv = BasicConv(out_channels, out_channels, 1)
def forward(self, x):
x = self.downsample_conv(x)
x0 = self.split_conv0(x)
x1 = self.split_conv1(x)
x1 = self.blocks_conv(x1)
x = torch.cat([x1, x0], dim=1)
x = self.concat_conv(x)
return x
class CSPDarkNet(nn.Module):
def __init__(self, layers):
super(CSPDarkNet, self).__init__()
self.inplanes = 32
self.conv1 = BasicConv(3, self.inplanes, kernel_size=3, stride=1)
self.feature_channels = [64, 128, 256, 512, 1024]
self.stages = nn.ModuleList([
Resblock_body(self.inplanes, self.feature_channels[0], layers[0], first=True),
Resblock_body(self.feature_channels[0], self.feature_channels[1], layers[1], first=False),
Resblock_body(self.feature_channels[1], self.feature_channels[2], layers[2], first=False),
Resblock_body(self.feature_channels[2], self.feature_channels[3], layers[3], first=False),
Resblock_body(self.feature_channels[3], self.feature_channels[4], layers[4], first=False)
])
self.num_features = 1
# 进行权值初始化
for m in self.modules():
if isinstance(m, nn.Conv2d):
n = m.kernel_size[0] * m.kernel_size[1] * m.out_channels
m.weight.data.normal_(0, math.sqrt(2. / n))
elif isinstance(m, nn.BatchNorm2d):
m.weight.data.fill_(1)
m.bias.data.zero_()
def forward(self, x):
x = self.conv1(x)
x = self.stages[0](x)
x = self.stages[1](x)
out3 = self.stages[2](x)
out4 = self.stages[3](out3)
out5 = self.stages[4](out4)
return out3, out4, out5
def darknet53(pretrained, **kwargs):
model = CSPDarkNet([1, 2, 8, 8, 4])
if pretrained:
if isinstance(pretrained, str):
model.load_state_dict(torch.load(pretrained))
else:
raise Exception("darknet request a pretrained path. got [{}]".format(pretrained))
return model
在上图中,除了CSPDarknet53和YOLO Head的结构之外,其余都是特征金字塔的结构。
在特征金字塔部分,YOLO V4 结合两种改进:SPP 结构 和 PANet 结构。
1、 SSP结构加在CSPdarknet53的最后一个特征层的卷积,在CSP darknet53的最后一个特征层进行3次DarknetConv2D_BN_Leaky卷积后, 分别利用 四个 不同尺度的最大池化 进行处理,最大池化的池化核大小如图所示, 分别为 13 x 13 , 9 x 9, 5 x 5, 1 x 1(无需处理)。
#---------------------------------------------------#
# SPP结构,利用不同大小的池化核进行池化
# 池化后堆叠
#---------------------------------------------------#
class SpatialPyramidPooling(nn.Module):
def __init__(self, pool_sizes=[5, 9, 13]):
super(SpatialPyramidPooling, self).__init__()
self.maxpools = nn.ModuleList([nn.MaxPool2d(pool_size, 1, pool_size//2) for pool_size in pool_sizes])
def forward(self, x):
features = [maxpool(x) for maxpool in self.maxpools[::-1]]
features = torch.cat(features + [x], dim=1)
return features
2、 PANet是 2018 年的一种实例分割算法。
上图是原始的PANet结构,从 先上后下再上 的曲线,看出具有一个非常重要的特点就是特征的反复提取。在(a)里面是传统的特征金字塔结构,完成从下到上的特征提取后, 还需要实现 (b)部分从上到下的特征提取。在YOLOV4中, 最后三个 有效特征层上 使用了 PANet结构。
实现代码:
#---------------------------------------------------#
# yolo_body
#---------------------------------------------------#
class YoloBody(nn.Module):
def __init__(self, config):
super(YoloBody, self).__init__()
self.config = config
# backbone
self.backbone = darknet53(None)
self.conv1 = make_three_conv([512,1024],1024)
self.SPP = SpatialPyramidPooling()
self.conv2 = make_three_conv([512,1024],2048)
self.upsample1 = Upsample(512,256)
self.conv_for_P4 = conv2d(512,256,1)
self.make_five_conv1 = make_five_conv([256, 512],512)
self.upsample2 = Upsample(256,128)
self.conv_for_P3 = conv2d(256,128,1)
self.make_five_conv2 = make_five_conv([128, 256],256)
# 3*(5+num_classes)=3*(5+20)=3*(4+1+20)=75
final_out_filter2 = len(config["yolo"]["anchors"][2]) * (5 + config["yolo"]["classes"])
self.yolo_head3 = yolo_head([256, final_out_filter2],128)
self.down_sample1 = conv2d(128,256,3,stride=2)
self.make_five_conv3 = make_five_conv([256, 512],512)
# 3*(5+num_classes)=3*(5+20)=3*(4+1+20)=75
final_out_filter1 = len(config["yolo"]["anchors"][1]) * (5 + config["yolo"]["classes"])
self.yolo_head2 = yolo_head([512, final_out_filter1],256)
self.down_sample2 = conv2d(256,512,3,stride=2)
self.make_five_conv4 = make_five_conv([512, 1024],1024)
# 3*(5+num_classes)=3*(5+20)=3*(4+1+20)=75
final_out_filter0 = len(config["yolo"]["anchors"][0]) * (5 + config["yolo"]["classes"])
self.yolo_head1 = yolo_head([1024, final_out_filter0],512)
def forward(self, x):
# backbone
x2, x1, x0 = self.backbone(x)
P5 = self.conv1(x0)
P5 = self.SPP(P5)
P5 = self.conv2(P5)
P5_upsample = self.upsample1(P5)
P4 = self.conv_for_P4(x1)
P4 = torch.cat([P4,P5_upsample],axis=1)
P4 = self.make_five_conv1(P4)
P4_upsample = self.upsample2(P4)
P3 = self.conv_for_P3(x2)
P3 = torch.cat([P3,P4_upsample],axis=1)
P3 = self.make_five_conv2(P3)
P3_downsample = self.down_sample1(P3)
P4 = torch.cat([P3_downsample,P4],axis=1)
P4 = self.make_five_conv3(P4)
P4_downsample = self.down_sample2(P4)
P5 = torch.cat([P4_downsample,P5],axis=1)
P5 = self.make_five_conv4(P5)
out2 = self.yolo_head3(P3)
out1 = self.yolo_head2(P4)
out0 = self.yolo_head1(P5)
return out0, out1, out2
3、YoLoHead 利用获得到的 特征进行预测
1. 这一部分和YOLO V3 中的过程是一样的。在特征利用部分, YOLO V4提取 多特征层进行目标检测, 这三个被提取的特征层 分别位于 中、中下、底层。在input的shape是 608 x 608 时,这三个特征层的shape 是 (76,76,256),(38,38,512),(19,19,1024)。
2. 输出层的shape分别是(19,19,75),(38,38,75),(76,76,75)。关于75的解释,75 = 3*(20+1+4),3是因为yolo v4针对每一个特征层存在3个先验框, 20是因为这个图的结构是 基于 voc数据集,因为voc数据集的类是20种,4是因为对先验框调整需要4个参数,1是表示置信度。比如,要换成coco数据集(含有80个不同的类别),那么“75”应该改变成 “3*(80+4+1)=255”
实现代码:
#---------------------------------------------------#
# 最后获得yolov4的输出
#---------------------------------------------------#
def yolo_head(filters_list, in_filters):
m = nn.Sequential(
conv2d(in_filters, filters_list[0], 3),
nn.Conv2d(filters_list[0], filters_list[1], 1),
)
return m
#---------------------------------------------------#
# yolo_body
#---------------------------------------------------#
class YoloBody(nn.Module):
def __init__(self, config):
super(YoloBody, self).__init__()
self.config = config
# backbone
self.backbone = darknet53(None)
self.conv1 = make_three_conv([512,1024],1024)
self.SPP = SpatialPyramidPooling()
self.conv2 = make_three_conv([512,1024],2048)
self.upsample1 = Upsample(512,256)
self.conv_for_P4 = conv2d(512,256,1)
self.make_five_conv1 = make_five_conv([256, 512],512)
self.upsample2 = Upsample(256,128)
self.conv_for_P3 = conv2d(256,128,1)
self.make_five_conv2 = make_five_conv([128, 256],256)
# 3*(5+num_classes)=3*(5+20)=3*(4+1+20)=75
final_out_filter2 = len(config["yolo"]["anchors"][2]) * (5 + config["yolo"]["classes"])
self.yolo_head3 = yolo_head([256, final_out_filter2],128)
self.down_sample1 = conv2d(128,256,3,stride=2)
self.make_five_conv3 = make_five_conv([256, 512],512)
# 3*(5+num_classes)=3*(5+20)=3*(4+1+20)=75
final_out_filter1 = len(config["yolo"]["anchors"][1]) * (5 + config["yolo"]["classes"])
self.yolo_head2 = yolo_head([512, final_out_filter1],256)
self.down_sample2 = conv2d(256,512,3,stride=2)
self.make_five_conv4 = make_five_conv([512, 1024],1024)
# 3*(5+num_classes)=3*(5+20)=3*(4+1+20)=75
final_out_filter0 = len(config["yolo"]["anchors"][0]) * (5 + config["yolo"]["classes"])
self.yolo_head1 = yolo_head([1024, final_out_filter0],512)
def forward(self, x):
# backbone
x2, x1, x0 = self.backbone(x)
P5 = self.conv1(x0)
P5 = self.SPP(P5)
P5 = self.conv2(P5)
P5_upsample = self.upsample1(P5)
P4 = self.conv_for_P4(x1)
P4 = torch.cat([P4,P5_upsample],axis=1)
P4 = self.make_five_conv1(P4)
P4_upsample = self.upsample2(P4)
P3 = self.conv_for_P3(x2)
P3 = torch.cat([P3,P4_upsample],axis=1)
P3 = self.make_five_conv2(P3)
P3_downsample = self.down_sample1(P3)
P4 = torch.cat([P3_downsample,P4],axis=1)
P4 = self.make_five_conv3(P4)
P4_downsample = self.down_sample2(P4)
P5 = torch.cat([P4_downsample,P5],axis=1)
P5 = self.make_five_conv4(P5)
out2 = self.yolo_head3(P3)
out1 = self.yolo_head2(P4)
out0 = self.yolo_head1(P5)
return out0, out1, out2
4、预测结果的解码
从第二步可以获得三个特征层的预测结果,shape分别为(N,19,19,75),(N,38,38,75),(N,76,76,75) 的数据,对应每个图被分成 19x 19、38 x 38、76 x 76 的网格上面 3个预测框的位置。但是,这个预测结果,同样的,不是最终预测框在图片上的位置,还需要经历解码才能完成。
在此,略提一些YOLO V3的预测原理,YOLO V3的 3个 特征层 分别将整幅图 分为 19x19、38 x 38、 76 x 76的网格,每个网络点负责某一个确定区域的检测。
在前已经提过,最后一个维度的“75”=(20+1+4)*3,这个(20+1+4)分别代表了 分类结果、 置信度、 (x_offset、y_offset、h和w)。
在YOLOV3的解码过程, 将每个网格点加上 它 对应的 x_offset 和 y_offset,加上之后对应的就是预测框的中心,然后结合 先验框 和 h 、w结合 计算出预测框的长和宽。
最后,需要对预测进行得分排序和非最大抑制筛选(NMS),才能得到最后的预测结果。这个原理,几乎所有目标检测都通用。但是该项目的处理方式,是对每一个类别进行判别:
实现代码如下,当调用yolo_eval时,就会对每个特征层进行解码:
class DecodeBox(nn.Module):
def __init__(self, anchors, num_classes, img_size):
super(DecodeBox, self).__init__()
self.anchors = anchors
self.num_anchors = len(anchors)
self.num_classes = num_classes
self.bbox_attrs = 5 + num_classes
self.img_size = img_size
def forward(self, input):
# input为bs,3*(1+4+num_classes),13,13
# 一共多少张图片
batch_size = input.size(0)
# 13,13
input_height = input.size(2)
input_width = input.size(3)
# 计算步长
# 每一个特征点对应原来的图片上多少个像素点
# 如果特征层为13x13的话,一个特征点就对应原来的图片上的32个像素点
# 416/13 = 32
stride_h = self.img_size[1] / input_height
stride_w = self.img_size[0] / input_width
# 把先验框的尺寸调整成特征层大小的形式
# 计算出先验框在特征层上对应的宽高
scaled_anchors = [(anchor_width / stride_w, anchor_height / stride_h) for anchor_width, anchor_height in self.anchors]
# bs,3*(5+num_classes),13,13 -> bs,3,13,13,(5+num_classes)
prediction = input.view(batch_size, self.num_anchors,
self.bbox_attrs, input_height, input_width).permute(0, 1, 3, 4, 2).contiguous()
# 先验框的中心位置的调整参数
x = torch.sigmoid(prediction[..., 0])
y = torch.sigmoid(prediction[..., 1])
# 先验框的宽高调整参数
w = prediction[..., 2] # Width
h = prediction[..., 3] # Height
# 获得置信度,是否有物体
conf = torch.sigmoid(prediction[..., 4])
# 种类置信度
pred_cls = torch.sigmoid(prediction[..., 5:]) # Cls pred.
FloatTensor = torch.cuda.FloatTensor if x.is_cuda else torch.FloatTensor
LongTensor = torch.cuda.LongTensor if x.is_cuda else torch.LongTensor
# 生成网格,先验框中心,网格左上角 batch_size,3,13,13
grid_x = torch.linspace(0, input_width - 1, input_width).repeat(input_width, 1).repeat(
batch_size * self.num_anchors, 1, 1).view(x.shape).type(FloatTensor)
grid_y = torch.linspace(0, input_height - 1, input_height).repeat(input_height, 1).t().repeat(
batch_size * self.num_anchors, 1, 1).view(y.shape).type(FloatTensor)
# 生成先验框的宽高
anchor_w = FloatTensor(scaled_anchors).index_select(1, LongTensor([0]))
anchor_h = FloatTensor(scaled_anchors).index_select(1, LongTensor([1]))
anchor_w = anchor_w.repeat(batch_size, 1).repeat(1, 1, input_height * input_width).view(w.shape)
anchor_h = anchor_h.repeat(batch_size, 1).repeat(1, 1, input_height * input_width).view(h.shape)
# 计算调整后的先验框中心与宽高
pred_boxes = FloatTensor(prediction[..., :4].shape)
pred_boxes[..., 0] = x.data + grid_x
pred_boxes[..., 1] = y.data + grid_y
pred_boxes[..., 2] = torch.exp(w.data) * anchor_w
pred_boxes[..., 3] = torch.exp(h.data) * anchor_h
# 用于将输出调整为相对于416x416的大小
_scale = torch.Tensor([stride_w, stride_h] * 2).type(FloatTensor)
output = torch.cat((pred_boxes.view(batch_size, -1, 4) * _scale,
conf.view(batch_size, -1, 1), pred_cls.view(batch_size, -1, self.num_classes)), -1)
return output.data
5、在原图上进行绘制
在第4步,可以获得预测框在原图上的真实位置,且都经过筛选,可以直接绘制在图片上。
a) Mosaic 数据增强
Mosaic利用了四张图片,根据论文说 其具备的巨大优点是: 丰富检测物体的背景, 并且在BN计算的时候会 一下子计算 四张图片的数据。实现思路如下:
1、每次读取四张照片
2、分别对四张照片进行反转、缩放、色域变化等,并且按照四个方向位置摆放好。
3、进行图片的组合和框的组合
具体步骤可以参考博客:https://blog.csdn.net/weixin_44791964/article/details/106214657
代码如下:
def rand(a=0, b=1):
return np.random.rand()*(b-a) + a
def merge_bboxes(bboxes, cutx, cuty):
merge_bbox = []
for i in range(len(bboxes)):
for box in bboxes[i]:
tmp_box = []
x1,y1,x2,y2 = box[0], box[1], box[2], box[3]
if i == 0:
if y1 > cuty or x1 > cutx:
continue
if y2 >= cuty and y1 <= cuty:
y2 = cuty
if y2-y1 < 5:
continue
if x2 >= cutx and x1 <= cutx:
x2 = cutx
if x2-x1 < 5:
continue
if i == 1:
if y2 < cuty or x1 > cutx:
continue
if y2 >= cuty and y1 <= cuty:
y1 = cuty
if y2-y1 < 5:
continue
if x2 >= cutx and x1 <= cutx:
x2 = cutx
if x2-x1 < 5:
continue
if i == 2:
if y2 < cuty or x2 < cutx:
continue
if y2 >= cuty and y1 <= cuty:
y1 = cuty
if y2-y1 < 5:
continue
if x2 >= cutx and x1 <= cutx:
x1 = cutx
if x2-x1 < 5:
continue
if i == 3:
if y1 > cuty or x2 < cutx:
continue
if y2 >= cuty and y1 <= cuty:
y2 = cuty
if y2-y1 < 5:
continue
if x2 >= cutx and x1 <= cutx:
x1 = cutx
if x2-x1 < 5:
continue
tmp_box.append(x1)
tmp_box.append(y1)
tmp_box.append(x2)
tmp_box.append(y2)
tmp_box.append(box[-1])
merge_bbox.append(tmp_box)
return merge_bbox
def get_random_data(annotation_line, input_shape, random=True, hue=.1, sat=1.5, val=1.5, proc_img=True):
'''random preprocessing for real-time data augmentation'''
h, w = input_shape
min_offset_x = 0.4
min_offset_y = 0.4
scale_low = 1-min(min_offset_x,min_offset_y)
scale_high = scale_low+0.2
image_datas = []
box_datas = []
index = 0
place_x = [0,0,int(w*min_offset_x),int(w*min_offset_x)]
place_y = [0,int(h*min_offset_y),int(w*min_offset_y),0]
for line in annotation_line:
# 每一行进行分割
line_content = line.split()
# 打开图片
image = Image.open(line_content[0])
image = image.convert("RGB")
# 图片的大小
iw, ih = image.size
# 保存框的位置
box = np.array([np.array(list(map(int,box.split(',')))) for box in line_content[1:]])
# image.save(str(index)+".jpg")
# 是否翻转图片
flip = rand()<.5
if flip and len(box)>0:
image = image.transpose(Image.FLIP_LEFT_RIGHT)
box[:, [0,2]] = iw - box[:, [2,0]]
# 对输入进来的图片进行缩放
new_ar = w/h
scale = rand(scale_low, scale_high)
if new_ar < 1:
nh = int(scale*h)
nw = int(nh*new_ar)
else:
nw = int(scale*w)
nh = int(nw/new_ar)
image = image.resize((nw,nh), Image.BICUBIC)
# 进行色域变换
hue = rand(-hue, hue)
sat = rand(1, sat) if rand()<.5 else 1/rand(1, sat)
val = rand(1, val) if rand()<.5 else 1/rand(1, val)
x = rgb_to_hsv(np.array(image)/255.)
x[..., 0] += hue
x[..., 0][x[..., 0]>1] -= 1
x[..., 0][x[..., 0]<0] += 1
x[..., 1] *= sat
x[..., 2] *= val
x[x>1] = 1
x[x<0] = 0
image = hsv_to_rgb(x)
image = Image.fromarray((image*255).astype(np.uint8))
# 将图片进行放置,分别对应四张分割图片的位置
dx = place_x[index]
dy = place_y[index]
new_image = Image.new('RGB', (w,h), (128,128,128))
new_image.paste(image, (dx, dy))
image_data = np.array(new_image)/255
# Image.fromarray((image_data*255).astype(np.uint8)).save(str(index)+"distort.jpg")
index = index + 1
box_data = []
# 对box进行重新处理
if len(box)>0:
np.random.shuffle(box)
box[:, [0,2]] = box[:, [0,2]]*nw/iw + dx
box[:, [1,3]] = box[:, [1,3]]*nh/ih + dy
box[:, 0:2][box[:, 0:2]<0] = 0
box[:, 2][box[:, 2]>w] = w
box[:, 3][box[:, 3]>h] = h
box_w = box[:, 2] - box[:, 0]
box_h = box[:, 3] - box[:, 1]
box = box[np.logical_and(box_w>1, box_h>1)]
box_data = np.zeros((len(box),5))
box_data[:len(box)] = box
image_datas.append(image_data)
box_datas.append(box_data)
img = Image.fromarray((image_data*255).astype(np.uint8))
for j in range(len(box_data)):
thickness = 3
left, top, right, bottom = box_data[j][0:4]
draw = ImageDraw.Draw(img)
for i in range(thickness):
draw.rectangle([left + i, top + i, right - i, bottom - i],outline=(255,255,255))
img.show()
# 将图片分割,放在一起
cutx = np.random.randint(int(w*min_offset_x), int(w*(1 - min_offset_x)))
cuty = np.random.randint(int(h*min_offset_y), int(h*(1 - min_offset_y)))
new_image = np.zeros([h,w,3])
new_image[:cuty, :cutx, :] = image_datas[0][:cuty, :cutx, :]
new_image[cuty:, :cutx, :] = image_datas[1][cuty:, :cutx, :]
new_image[cuty:, cutx:, :] = image_datas[2][cuty:, cutx:, :]
new_image[:cuty, cutx:, :] = image_datas[3][:cuty, cutx:, :]
# 对框进行进一步的处理
new_boxes = merge_bboxes(box_datas, cutx, cuty)
return new_image, new_boxes
b)Label Smoothing平滑(搬运)
标签平滑具体公式如下:
new_onehot_labels = onehot_labels * (1 - label_smoothing) + label_smoothing / num_classes
当label_smoothing=0.01时,公式变成如下:
new_onehot_labels = y * (1 - 0.01) + 0.01 / num_classes
例如,原始的onehot_labels为0、1时,在平滑后会变成0.005、0.995(二分类,num_classes=2),which means 对分类准确做一点惩罚,让模型不可以分类地太准确,太准确会容易导致过拟合问题。
c)CIOU
CIOU将目标与anchor之间的距离、重叠率、尺度以及惩罚项都考虑进去。
d)学习率余弦退火衰减
余弦退火算法:学习率会先上升再下降,上升的时候使用线性上升,下降的时候模拟cos函数下降,执行多次。
pytorch有直接的实现函数,可以直接调用
lr_scheduler = optim.lr_scheduler.CosineAnnealingLR(optimizer, T_max=5, eta_min=1e-5)
a) 计算loss所需参数
loss的计算过程,实际是y_pre和y_true之间的对比:y_pre 是一幅图像经过网络之后的输出,内部有三个特征层的内容;其需要解码才能够在图上作画; y_true 是一个真实图像中,他的每个 真实框 对应的 19 x 19 , 38 x 38 , 76 x 76 网格上的偏移位置、长宽与种类,需要编码才能和y_pred 的结构一致。
更深一步理解, y_pre 和 y_true 的shape是
(batch_size,19,19,3,25); (batch_size,38,38,3,25);(batch_size,76,76,3,25);
b) y_pre
网络最后输出的内容是三个有效特征层每个网格点对应的预测框及其种类,三个特征层分别对应着图片被resize成不同尺寸的网格后,每个网格点上 三个先验框 对应的位置、 置信度 及其种类。
对于输出的 y1 、y2、 y3而言,[...,:2]指的是相对于每个网格点的偏移量,[...,2:4]指的是宽和高,[...,4:5]指的是该框的置信度,[...,5:]指的是每个种类的预测概率。
c) y_true
y_true就是一个真实图像中,它的每个真实框对应的(19,19)、(38,38)、(76,76)网格上的偏移位置、长宽与种类。其仍需要编码才能与y_pred的结构一致
d)loss的计算过程
loss值需要对 三个特征层进行处理,而不是把 y_pre和y_true简单的相减。
这里以最小的特征层 19 x 19为例。
代码如下:
#---------------------------------------------------#
# 平滑标签
#---------------------------------------------------#
def smooth_labels(y_true, label_smoothing,num_classes):
return y_true * (1.0 - label_smoothing) + label_smoothing / num_classes
def box_ciou(b1, b2):
"""
输入为:
----------
b1: tensor, shape=(batch, feat_w, feat_h, anchor_num, 4), xywh
b2: tensor, shape=(batch, feat_w, feat_h, anchor_num, 4), xywh
返回为:
-------
ciou: tensor, shape=(batch, feat_w, feat_h, anchor_num, 1)
"""
# 求出预测框左上角右下角
b1_xy = b1[..., :2]
b1_wh = b1[..., 2:4]
b1_wh_half = b1_wh/2.
b1_mins = b1_xy - b1_wh_half
b1_maxes = b1_xy + b1_wh_half
# 求出真实框左上角右下角
b2_xy = b2[..., :2]
b2_wh = b2[..., 2:4]
b2_wh_half = b2_wh/2.
b2_mins = b2_xy - b2_wh_half
b2_maxes = b2_xy + b2_wh_half
# 求真实框和预测框所有的iou
intersect_mins = torch.max(b1_mins, b2_mins)
intersect_maxes = torch.min(b1_maxes, b2_maxes)
intersect_wh = torch.max(intersect_maxes - intersect_mins, torch.zeros_like(intersect_maxes))
intersect_area = intersect_wh[..., 0] * intersect_wh[..., 1]
b1_area = b1_wh[..., 0] * b1_wh[..., 1]
b2_area = b2_wh[..., 0] * b2_wh[..., 1]
union_area = b1_area + b2_area - intersect_area
iou = intersect_area / (union_area + 1e-6)
# 计算中心的差距
center_distance = torch.sum(torch.pow((b1_xy - b2_xy), 2), axis=-1)
# 找到包裹两个框的最小框的左上角和右下角
enclose_mins = torch.min(b1_mins, b2_mins)
enclose_maxes = torch.max(b1_maxes, b2_maxes)
enclose_wh = torch.max(enclose_maxes - enclose_mins, torch.zeros_like(intersect_maxes))
# 计算对角线距离
enclose_diagonal = torch.sum(torch.pow(enclose_wh,2), axis=-1)
ciou = iou - 1.0 * (center_distance) / (enclose_diagonal + 1e-7)
v = (4 / (math.pi ** 2)) * torch.pow((torch.atan(b1_wh[..., 0]/b1_wh[..., 1]) - torch.atan(b2_wh[..., 0]/b2_wh[..., 1])), 2)
alpha = v / (1.0 - iou + v)
ciou = ciou - alpha * v
return ciou
def clip_by_tensor(t,t_min,t_max):
t=t.float()
result = (t >= t_min).float() * t + (t < t_min).float() * t_min
result = (result <= t_max).float() * result + (result > t_max).float() * t_max
return result
def MSELoss(pred,target):
return (pred-target)**2
def BCELoss(pred,target):
epsilon = 1e-7
pred = clip_by_tensor(pred, epsilon, 1.0 - epsilon)
output = -target * torch.log(pred) - (1.0 - target) * torch.log(1.0 - pred)
return output
class YOLOLoss(nn.Module):
def __init__(self, anchors, num_classes, img_size, label_smooth=0, cuda=True):
super(YOLOLoss, self).__init__()
self.anchors = anchors
self.num_anchors = len(anchors)
self.num_classes = num_classes
self.bbox_attrs = 5 + num_classes
self.img_size = img_size
self.label_smooth = label_smooth
self.ignore_threshold = 0.5
self.lambda_conf = 1.0
self.lambda_cls = 1.0
self.lambda_loc = 1.0
self.cuda = cuda
def forward(self, input, targets=None):
# input为bs,3*(5+num_classes),13,13
# 一共多少张图片
bs = input.size(0)
# 特征层的高
in_h = input.size(2)
# 特征层的宽
in_w = input.size(3)
# 计算步长
# 每一个特征点对应原来的图片上多少个像素点
# 如果特征层为13x13的话,一个特征点就对应原来的图片上的32个像素点
stride_h = self.img_size[1] / in_h
stride_w = self.img_size[0] / in_w
# 把先验框的尺寸调整成特征层大小的形式
# 计算出先验框在特征层上对应的宽高
scaled_anchors = [(a_w / stride_w, a_h / stride_h) for a_w, a_h in self.anchors]
# bs,3*(5+num_classes),13,13 -> bs,3,13,13,(5+num_classes)
prediction = input.view(bs, int(self.num_anchors/3),
self.bbox_attrs, in_h, in_w).permute(0, 1, 3, 4, 2).contiguous()
# 对prediction预测进行调整
conf = torch.sigmoid(prediction[..., 4]) # Conf
pred_cls = torch.sigmoid(prediction[..., 5:]) # Cls pred.
# 找到哪些先验框内部包含物体
mask, noobj_mask, t_box, tconf, tcls, box_loss_scale_x, box_loss_scale_y = self.get_target(targets, scaled_anchors,in_w, in_h,self.ignore_threshold)
noobj_mask, pred_boxes_for_ciou = self.get_ignore(prediction, targets, scaled_anchors, in_w, in_h, noobj_mask)
if self.cuda:
mask, noobj_mask = mask.cuda(), noobj_mask.cuda()
box_loss_scale_x, box_loss_scale_y= box_loss_scale_x.cuda(), box_loss_scale_y.cuda()
tconf, tcls = tconf.cuda(), tcls.cuda()
pred_boxes_for_ciou = pred_boxes_for_ciou.cuda()
t_box = t_box.cuda()
box_loss_scale = 2-box_loss_scale_x*box_loss_scale_y
# losses.
ciou = (1 - box_ciou( pred_boxes_for_ciou[mask.bool()], t_box[mask.bool()]))* box_loss_scale[mask.bool()]
loss_loc = torch.sum(ciou / bs)
loss_conf = torch.sum(BCELoss(conf, mask) * mask / bs) + \
torch.sum(BCELoss(conf, mask) * noobj_mask / bs)
# print(smooth_labels(tcls[mask == 1],self.label_smooth,self.num_classes))
loss_cls = torch.sum(BCELoss(pred_cls[mask == 1], smooth_labels(tcls[mask == 1],self.label_smooth,self.num_classes))/bs)
# print(loss_loc,loss_conf,loss_cls)
loss = loss_conf * self.lambda_conf + loss_cls * self.lambda_cls + loss_loc * self.lambda_loc
return loss, loss_conf.item(), loss_cls.item(), loss_loc.item()
def get_target(self, target, anchors, in_w, in_h, ignore_threshold):
# 计算一共有多少张图片
bs = len(target)
# 获得先验框
anchor_index = [[0,1,2],[3,4,5],[6,7,8]][[13,26,52].index(in_w)]
subtract_index = [0,3,6][[13,26,52].index(in_w)]
# 创建全是0或者全是1的阵列
mask = torch.zeros(bs, int(self.num_anchors/3), in_h, in_w, requires_grad=False)
noobj_mask = torch.ones(bs, int(self.num_anchors/3), in_h, in_w, requires_grad=False)
tx = torch.zeros(bs, int(self.num_anchors/3), in_h, in_w, requires_grad=False)
ty = torch.zeros(bs, int(self.num_anchors/3), in_h, in_w, requires_grad=False)
tw = torch.zeros(bs, int(self.num_anchors/3), in_h, in_w, requires_grad=False)
th = torch.zeros(bs, int(self.num_anchors/3), in_h, in_w, requires_grad=False)
t_box = torch.zeros(bs, int(self.num_anchors/3), in_h, in_w, 4, requires_grad=False)
tconf = torch.zeros(bs, int(self.num_anchors/3), in_h, in_w, requires_grad=False)
tcls = torch.zeros(bs, int(self.num_anchors/3), in_h, in_w, self.num_classes, requires_grad=False)
box_loss_scale_x = torch.zeros(bs, int(self.num_anchors/3), in_h, in_w, requires_grad=False)
box_loss_scale_y = torch.zeros(bs, int(self.num_anchors/3), in_h, in_w, requires_grad=False)
for b in range(bs):
for t in range(target[b].shape[0]):
# 计算出在特征层上的点位
gx = target[b][t, 0] * in_w
gy = target[b][t, 1] * in_h
gw = target[b][t, 2] * in_w
gh = target[b][t, 3] * in_h
# 计算出属于哪个网格
gi = int(gx)
gj = int(gy)
# 计算真实框的位置
gt_box = torch.FloatTensor(np.array([0, 0, gw, gh])).unsqueeze(0)
# 计算出所有先验框的位置
anchor_shapes = torch.FloatTensor(np.concatenate((np.zeros((self.num_anchors, 2)),
np.array(anchors)), 1))
# 计算重合程度
anch_ious = bbox_iou(gt_box, anchor_shapes)
# Find the best matching anchor box
best_n = np.argmax(anch_ious)
if best_n not in anchor_index:
continue
# Masks
if (gj < in_h) and (gi < in_w):
best_n = best_n - subtract_index
# 判定哪些先验框内部真实的存在物体
noobj_mask[b, best_n, gj, gi] = 0
mask[b, best_n, gj, gi] = 1
# 计算先验框中心调整参数
tx[b, best_n, gj, gi] = gx
ty[b, best_n, gj, gi] = gy
# 计算先验框宽高调整参数
tw[b, best_n, gj, gi] = gw
th[b, best_n, gj, gi] = gh
# 用于获得xywh的比例
box_loss_scale_x[b, best_n, gj, gi] = target[b][t, 2]
box_loss_scale_y[b, best_n, gj, gi] = target[b][t, 3]
# 物体置信度
tconf[b, best_n, gj, gi] = 1
# 种类
tcls[b, best_n, gj, gi, int(target[b][t, 4])] = 1
else:
print('Step {0} out of bound'.format(b))
print('gj: {0}, height: {1} | gi: {2}, width: {3}'.format(gj, in_h, gi, in_w))
continue
t_box[...,0] = tx
t_box[...,1] = ty
t_box[...,2] = tw
t_box[...,3] = th
return mask, noobj_mask, t_box, tconf, tcls, box_loss_scale_x, box_loss_scale_y
def get_ignore(self,prediction,target,scaled_anchors,in_w, in_h,noobj_mask):
bs = len(target)
anchor_index = [[0,1,2],[3,4,5],[6,7,8]][[13,26,52].index(in_w)]
scaled_anchors = np.array(scaled_anchors)[anchor_index]
# 先验框的中心位置的调整参数
x = torch.sigmoid(prediction[..., 0])
y = torch.sigmoid(prediction[..., 1])
# 先验框的宽高调整参数
w = prediction[..., 2] # Width
h = prediction[..., 3] # Height
FloatTensor = torch.cuda.FloatTensor if x.is_cuda else torch.FloatTensor
LongTensor = torch.cuda.LongTensor if x.is_cuda else torch.LongTensor
# 生成网格,先验框中心,网格左上角
grid_x = torch.linspace(0, in_w - 1, in_w).repeat(in_w, 1).repeat(
int(bs*self.num_anchors/3), 1, 1).view(x.shape).type(FloatTensor)
grid_y = torch.linspace(0, in_h - 1, in_h).repeat(in_h, 1).t().repeat(
int(bs*self.num_anchors/3), 1, 1).view(y.shape).type(FloatTensor)
# 生成先验框的宽高
anchor_w = FloatTensor(scaled_anchors).index_select(1, LongTensor([0]))
anchor_h = FloatTensor(scaled_anchors).index_select(1, LongTensor([1]))
anchor_w = anchor_w.repeat(bs, 1).repeat(1, 1, in_h * in_w).view(w.shape)
anchor_h = anchor_h.repeat(bs, 1).repeat(1, 1, in_h * in_w).view(h.shape)
# 计算调整后的先验框中心与宽高
pred_boxes = FloatTensor(prediction[..., :4].shape)
pred_boxes[..., 0] = x + grid_x
pred_boxes[..., 1] = y + grid_y
pred_boxes[..., 2] = torch.exp(w) * anchor_w
pred_boxes[..., 3] = torch.exp(h) * anchor_h
for i in range(bs):
pred_boxes_for_ignore = pred_boxes[i]
pred_boxes_for_ignore = pred_boxes_for_ignore.view(-1, 4)
for t in range(target[i].shape[0]):
gx = target[i][t, 0] * in_w
gy = target[i][t, 1] * in_h
gw = target[i][t, 2] * in_w
gh = target[i][t, 3] * in_h
gt_box = torch.FloatTensor(np.array([gx, gy, gw, gh])).unsqueeze(0).type(FloatTensor)
anch_ious = bbox_iou(gt_box, pred_boxes_for_ignore, x1y1x2y2=False)
anch_ious = anch_ious.view(pred_boxes[i].size()[:3])
noobj_mask[i][anch_ious>self.ignore_threshold] = 0
return noobj_mask, pred_boxes
总loss是三个loss的和,三个loss分别是: