前言
代码:https://github.com/2012Netsky/pytorch_cnn/blob/main/4_time_series_bikes.ipynb
参考:https://blog.csdn.net/nan355655600/article/details/106246625
一、yoloV3
from collections import OrderedDict
import torch
import torch.nn as nn
from nets.darknet import darknet53
def conv2d(filter_in, filter_out, kernel_size):
pad = (kernel_size - 1) // 2 if kernel_size else 0
return nn.Sequential(OrderedDict([
("conv", nn.Conv2d(filter_in, filter_out, kernel_size=kernel_size, stride=1, padding=pad, bias=False)),
("bn", nn.BatchNorm2d(filter_out)),
("relu", nn.LeakyReLU(0.1)),
]))
#------------------------------------------------------------------------#
# make_last_layers里面一共有七个卷积,前五个用于提取特征。
# 后两个用于获得yolo网络的预测结果
#------------------------------------------------------------------------#
def make_last_layers(filters_list, in_filters, out_filter):
m = nn.Sequential(
conv2d(in_filters, filters_list[0], 1),
conv2d(filters_list[0], filters_list[1], 3),
conv2d(filters_list[1], filters_list[0], 1),
conv2d(filters_list[0], filters_list[1], 3),
conv2d(filters_list[1], filters_list[0], 1),
conv2d(filters_list[0], filters_list[1], 3),
nn.Conv2d(filters_list[1], out_filter, kernel_size=1, stride=1, padding=0, bias=True)
)
# torch.nn.Sequential是一个Sequential容器,模块将按照构造函数中传递的顺序添加到模块中。
return m
# 我们在定义自已的网络的时候,需要继承nn.Module类,并重新实现构造函数__init__构造函数和forward这两个方法
class YoloBody(nn.Module):
def __init__(self, anchors_mask, num_classes, pretrained = False):
super(YoloBody, self).__init__() #调用父类的构造函数
#---------------------------------------------------#
# 生成darknet53的主干模型
# 获得三个有效特征层,他们的shape分别是:
# 52,52,256
# 26,26,512
# 13,13,1024
#---------------------------------------------------#
self.backbone = darknet53() # 自定义的model(继承nn.Module的model)
if pretrained:
self.backbone.load_state_dict(torch.load("model_data/darknet53_backbone_weights.pth"))
# Pytorch:模型的保存与加载 torch.save()、torch.load()、torch.nn.Module.load_state_dict() Pytorch
# 保存和加载模型后缀:.pt 和.pth
#---------------------------------------------------#
# out_filters : [64, 128, 256, 512, 1024]
#---------------------------------------------------#
out_filters = self.backbone.layers_out_filters # 成员变量
#------------------------------------------------------------------------#
# 计算yolo_head的输出通道数,对于voc数据集而言
# final_out_filter0 = final_out_filter1 = final_out_filter2 = 75
#------------------------------------------------------------------------#
self.last_layer0 = make_last_layers([512, 1024], out_filters[-1], len(anchors_mask[0]) * (num_classes + 5))
self.last_layer1_conv = conv2d(512, 256, 1)
self.last_layer1_upsample = nn.Upsample(scale_factor=2, mode='nearest') # pytorch torch.nn 实现上采样——nn.Upsample
self.last_layer1 = make_last_layers([256, 512], out_filters[-2] + 256, len(anchors_mask[1]) * (num_classes + 5))
self.last_layer2_conv = conv2d(256, 128, 1)
self.last_layer2_upsample = nn.Upsample(scale_factor=2, mode='nearest')
self.last_layer2 = make_last_layers([128, 256], out_filters[-3] + 128, len(anchors_mask[2]) * (num_classes + 5))
def forward(self, x):
#---------------------------------------------------#
# 获得三个有效特征层,他们的shape分别是:
# 52,52,256;26,26,512;13,13,1024
#---------------------------------------------------#
x2, x1, x0 = self.backbone(x)
#---------------------------------------------------#
# 第一个特征层
# out0 = (batch_size,255,13,13)
#---------------------------------------------------#
# 13,13,1024 -> 13,13,512 -> 13,13,1024 -> 13,13,512 -> 13,13,1024 -> 13,13,512
out0_branch = self.last_layer0[:5](x0)
out0 = self.last_layer0[5:](out0_branch)
# 13,13,512 -> 13,13,256 -> 26,26,256
x1_in = self.last_layer1_conv(out0_branch)
x1_in = self.last_layer1_upsample(x1_in)
# 26,26,256 + 26,26,512 -> 26,26,768
x1_in = torch.cat([x1_in, x1], 1)
#---------------------------------------------------#
# 第二个特征层
# out1 = (batch_size,255,26,26)
#---------------------------------------------------#
# 26,26,768 -> 26,26,256 -> 26,26,512 -> 26,26,256 -> 26,26,512 -> 26,26,256
out1_branch = self.last_layer1[:5](x1_in)
out1 = self.last_layer1[5:](out1_branch)
# 26,26,256 -> 26,26,128 -> 52,52,128
x2_in = self.last_layer2_conv(out1_branch)
x2_in = self.last_layer2_upsample(x2_in)
# 52,52,128 + 52,52,256 -> 52,52,384
x2_in = torch.cat([x2_in, x2], 1)
#---------------------------------------------------#
# 第一个特征层
# out3 = (batch_size,255,52,52)
#---------------------------------------------------#
# 52,52,384 -> 52,52,128 -> 52,52,256 -> 52,52,128 -> 52,52,256 -> 52,52,128
out2 = self.last_layer2(x2_in)
return out0, out1, out2
二、yoloV4
import colorsys
import os
import time
import numpy as np
import tensorflow as tf
from PIL import ImageDraw, ImageFont
from tensorflow.keras.layers import Input, Lambda
from tensorflow.keras.models import Model
from nets.yolo import yolo_body
from utils.utils import (cvtColor, get_anchors, get_classes, preprocess_input,
resize_image)
from utils.utils_bbox import DecodeBox
class YOLO(object):
_defaults = {
#--------------------------------------------------------------------------#
# 使用自己训练好的模型进行预测一定要修改model_path和classes_path!
# model_path指向logs文件夹下的权值文件,classes_path指向model_data下的txt
#
# 训练好后logs文件夹下存在多个权值文件,选择验证集损失较低的即可。
# 验证集损失较低不代表mAP较高,仅代表该权值在验证集上泛化性能较好。
# 如果出现shape不匹配,同时要注意训练时的model_path和classes_path参数的修改
#--------------------------------------------------------------------------#
"model_path" : 'model_data/yolov4_mobilenet_v1_voc.h5',
"classes_path" : 'model_data/voc_classes.txt',
#---------------------------------------------------------------------#
# anchors_path代表先验框对应的txt文件,一般不修改。
# anchors_mask用于帮助代码找到对应的先验框,一般不修改。
#---------------------------------------------------------------------#
"anchors_path" : 'model_data/yolo_anchors.txt',
"anchors_mask" : [[6, 7, 8], [3, 4, 5], [0, 1, 2]],
#---------------------------------------------------------------------#
# 输入图片的大小,必须为32的倍数。
#---------------------------------------------------------------------#
"input_shape" : [416, 416],
#---------------------------------------------------------------------#
# 目标检测网络所使用的主干
# 注意权值和主干的匹配
#---------------------------------------------------------------------#
"backbone" : 'mobilenetv1',
#---------------------------------------------------------------------#
# 通道的缩放系数
#---------------------------------------------------------------------#
"alpha" : 1,
#---------------------------------------------------------------------#
# 只有得分大于置信度的预测框会被保留下来
#---------------------------------------------------------------------#
"confidence" : 0.5,
#---------------------------------------------------------------------#
# 非极大抑制所用到的nms_iou大小
#---------------------------------------------------------------------#
"nms_iou" : 0.3,
"max_boxes" : 100,
#---------------------------------------------------------------------#
# 该变量用于控制是否使用letterbox_image对输入图像进行不失真的resize,
# 在多次测试后,发现关闭letterbox_image直接resize的效果更好
#---------------------------------------------------------------------#
"letterbox_image" : True,
}
@classmethod
def get_defaults(cls, n):
if n in cls._defaults:
return cls._defaults[n]
else:
return "Unrecognized attribute name '" + n + "'"
#---------------------------------------------------#
# 初始化yolo
#---------------------------------------------------#
def __init__(self, **kwargs):
self.__dict__.update(self._defaults)
for name, value in kwargs.items():
setattr(self, name, value)
#---------------------------------------------------#
# 获得种类和先验框的数量
#---------------------------------------------------#
self.class_names, self.num_classes = get_classes(self.classes_path)
self.anchors, self.num_anchors = get_anchors(self.anchors_path)
#---------------------------------------------------#
# 画框设置不同的颜色
#---------------------------------------------------#
hsv_tuples = [(x / self.num_classes, 1., 1.) for x in range(self.num_classes)]
self.colors = list(map(lambda x: colorsys.hsv_to_rgb(*x), hsv_tuples))
self.colors = list(map(lambda x: (int(x[0] * 255), int(x[1] * 255), int(x[2] * 255)), self.colors))
self.generate()
#---------------------------------------------------#
# 载入模型
#---------------------------------------------------#
def generate(self):
model_path = os.path.expanduser(self.model_path)
assert model_path.endswith('.h5'), 'Keras model or weights must be a .h5 file.'
self.model = yolo_body([None, None, 3], self.anchors_mask, self.num_classes, self.backbone, self.alpha)
self.model.load_weights(self.model_path)
print('{} model, anchors, and classes loaded.'.format(model_path))
#---------------------------------------------------------#
# 在DecodeBox函数中,我们会对预测结果进行后处理
# 后处理的内容包括,解码、非极大抑制、门限筛选等
#---------------------------------------------------------#
self.input_image_shape = Input([2,],batch_size=1)
inputs = [*self.model.output, self.input_image_shape]
outputs = Lambda(
DecodeBox,
output_shape = (1,),
name = 'yolo_eval',
arguments = {
'anchors' : self.anchors,
'num_classes' : self.num_classes,
'input_shape' : self.input_shape,
'anchor_mask' : self.anchors_mask,
'confidence' : self.confidence,
'nms_iou' : self.nms_iou,
'max_boxes' : self.max_boxes,
'letterbox_image' : self.letterbox_image
}
)(inputs)
self.yolo_model = Model([self.model.input, self.input_image_shape], outputs)
@tf.function
def get_pred(self, image_data, input_image_shape):
out_boxes, out_scores, out_classes = self.yolo_model([image_data, input_image_shape], training=False)
return out_boxes, out_scores, out_classes
#---------------------------------------------------#
# 检测图片
#---------------------------------------------------#
def detect_image(self, image, crop = False, count = False):
#---------------------------------------------------------#
# 在这里将图像转换成RGB图像,防止灰度图在预测时报错。
# 代码仅仅支持RGB图像的预测,所有其它类型的图像都会转化成RGB
#---------------------------------------------------------#
image = cvtColor(image)
#---------------------------------------------------------#
# 给图像增加灰条,实现不失真的resize
# 也可以直接resize进行识别
#---------------------------------------------------------#
image_data = resize_image(image, (self.input_shape[1], self.input_shape[0]), self.letterbox_image)
#---------------------------------------------------------#
# 添加上batch_size维度,并进行归一化
#---------------------------------------------------------#
image_data = np.expand_dims(preprocess_input(np.array(image_data, dtype='float32')), 0)
#---------------------------------------------------------#
# 将图像输入网络当中进行预测!
#---------------------------------------------------------#
input_image_shape = np.expand_dims(np.array([image.size[1], image.size[0]], dtype='float32'), 0)
out_boxes, out_scores, out_classes = self.get_pred(image_data, input_image_shape)
print('Found {} boxes for {}'.format(len(out_boxes), 'img'))
#---------------------------------------------------------#
# 设置字体与边框厚度
#---------------------------------------------------------#
font = ImageFont.truetype(font='model_data/simhei.ttf', size=np.floor(3e-2 * image.size[1] + 0.5).astype('int32'))
thickness = int(max((image.size[0] + image.size[1]) // np.mean(self.input_shape), 1))
#---------------------------------------------------------#
# 计数
#---------------------------------------------------------#
if count:
print("top_label:", out_classes)
classes_nums = np.zeros([self.num_classes])
for i in range(self.num_classes):
num = np.sum(out_classes == i)
if num > 0:
print(self.class_names[i], " : ", num)
classes_nums[i] = num
print("classes_nums:", classes_nums)
#---------------------------------------------------------#
# 是否进行目标的裁剪
#---------------------------------------------------------#
if crop:
for i, c in list(enumerate(out_boxes)):
top, left, bottom, right = out_boxes[i]
top = max(0, np.floor(top).astype('int32'))
left = max(0, np.floor(left).astype('int32'))
bottom = min(image.size[1], np.floor(bottom).astype('int32'))
right = min(image.size[0], np.floor(right).astype('int32'))
dir_save_path = "img_crop"
if not os.path.exists(dir_save_path):
os.makedirs(dir_save_path)
crop_image = image.crop([left, top, right, bottom])
crop_image.save(os.path.join(dir_save_path, "crop_" + str(i) + ".png"), quality=95, subsampling=0)
print("save crop_" + str(i) + ".png to " + dir_save_path)
#---------------------------------------------------------#
# 图像绘制
#---------------------------------------------------------#
for i, c in list(enumerate(out_classes)):
predicted_class = self.class_names[int(c)]
box = out_boxes[i]
score = out_scores[i]
top, left, bottom, right = box
top = max(0, np.floor(top).astype('int32'))
left = max(0, np.floor(left).astype('int32'))
bottom = min(image.size[1], np.floor(bottom).astype('int32'))
right = min(image.size[0], np.floor(right).astype('int32'))
label = '{} {:.2f}'.format(predicted_class, score)
draw = ImageDraw.Draw(image)
label_size = draw.textsize(label, font)
label = label.encode('utf-8')
print(label, top, left, bottom, right)
if top - label_size[1] >= 0:
text_origin = np.array([left, top - label_size[1]])
else:
text_origin = np.array([left, top + 1])
for i in range(thickness):
draw.rectangle([left + i, top + i, right - i, bottom - i], outline=self.colors[c])
draw.rectangle([tuple(text_origin), tuple(text_origin + label_size)], fill=self.colors[c])
draw.text(text_origin, str(label,'UTF-8'), fill=(0, 0, 0), font=font)
del draw
return image
def get_FPS(self, image, test_interval):
#---------------------------------------------------------#
# 在这里将图像转换成RGB图像,防止灰度图在预测时报错。
# 代码仅仅支持RGB图像的预测,所有其它类型的图像都会转化成RGB
#---------------------------------------------------------#
image = cvtColor(image)
#---------------------------------------------------------#
# 给图像增加灰条,实现不失真的resize
# 也可以直接resize进行识别
#---------------------------------------------------------#
image_data = resize_image(image, (self.input_shape[1], self.input_shape[0]), self.letterbox_image)
#---------------------------------------------------------#
# 添加上batch_size维度,并进行归一化
#---------------------------------------------------------#
image_data = np.expand_dims(preprocess_input(np.array(image_data, dtype='float32')), 0)
#---------------------------------------------------------#
# 将图像输入网络当中进行预测!
#---------------------------------------------------------#
input_image_shape = np.expand_dims(np.array([image.size[1], image.size[0]], dtype='float32'), 0)
out_boxes, out_scores, out_classes = self.get_pred(image_data, input_image_shape)
t1 = time.time()
for _ in range(test_interval):
out_boxes, out_scores, out_classes = self.get_pred(image_data, input_image_shape)
t2 = time.time()
tact_time = (t2 - t1) / test_interval
return tact_time
def detect_heatmap(self, image, heatmap_save_path):
import cv2
import matplotlib.pyplot as plt
def sigmoid(x):
y = 1.0 / (1.0 + np.exp(-x))
return y
#---------------------------------------------------------#
# 在这里将图像转换成RGB图像,防止灰度图在预测时报错。
# 代码仅仅支持RGB图像的预测,所有其它类型的图像都会转化成RGB
#---------------------------------------------------------#
image = cvtColor(image)
#---------------------------------------------------------#
# 给图像增加灰条,实现不失真的resize
# 也可以直接resize进行识别
#---------------------------------------------------------#
image_data = resize_image(image, (self.input_shape[1], self.input_shape[0]), self.letterbox_image)
#---------------------------------------------------------#
# 添加上batch_size维度,并进行归一化
#---------------------------------------------------------#
image_data = np.expand_dims(preprocess_input(np.array(image_data, dtype='float32')), 0)
output = self.model.predict(image_data)
plt.imshow(image, alpha=1)
plt.axis('off')
mask = np.zeros((image.size[1], image.size[0]))
for sub_output in output:
b, h, w, c = np.shape(sub_output)
sub_output = np.reshape(sub_output, [b, h, w, 3, -1])[0]
score = np.max(sigmoid(sub_output[..., 4]), -1)
score = cv2.resize(score, (image.size[0], image.size[1]))
normed_score = (score * 255).astype('uint8')
mask = np.maximum(mask, normed_score)
plt.imshow(mask, alpha=0.5, interpolation='nearest', cmap="jet")
plt.axis('off')
plt.subplots_adjust(top=1, bottom=0, right=1, left=0, hspace=0, wspace=0)
plt.margins(0, 0)
plt.savefig(heatmap_save_path, dpi=200, bbox_inches='tight', pad_inches = -0.1)
print("Save to the " + heatmap_save_path)
plt.show()
#---------------------------------------------------#
# 检测图片
#---------------------------------------------------#
def get_map_txt(self, image_id, image, class_names, map_out_path):
f = open(os.path.join(map_out_path, "detection-results/"+image_id+".txt"),"w")
#---------------------------------------------------------#
# 在这里将图像转换成RGB图像,防止灰度图在预测时报错。
#---------------------------------------------------------#
image = cvtColor(image)
#---------------------------------------------------------#
# 给图像增加灰条,实现不失真的resize
# 也可以直接resize进行识别
#---------------------------------------------------------#
image_data = resize_image(image, (self.input_shape[1], self.input_shape[0]), self.letterbox_image)
#---------------------------------------------------------#
# 添加上batch_size维度,并进行归一化
#---------------------------------------------------------#
image_data = np.expand_dims(preprocess_input(np.array(image_data, dtype='float32')), 0)
#---------------------------------------------------------#
# 将图像输入网络当中进行预测!
#---------------------------------------------------------#
input_image_shape = np.expand_dims(np.array([image.size[1], image.size[0]], dtype='float32'), 0)
out_boxes, out_scores, out_classes = self.get_pred(image_data, input_image_shape)
for i, c in enumerate(out_classes):
predicted_class = self.class_names[int(c)]
try:
score = str(out_scores[i].numpy())
except:
score = str(out_scores[i])
top, left, bottom, right = out_boxes[i]
if predicted_class not in class_names:
continue
f.write("%s %s %s %s %s %s\n" % (predicted_class, score[:6], str(int(left)), str(int(top)), str(int(right)),str(int(bottom))))
f.close()
return