虚拟数据集录制后用于训练操作
1.处理物体信息 把空格改成“，” gt含义obj_id, det_id, kf_id, min_x, min_y, max_x, max_y
import os
import numpy as np
import pandas as pd
import os.path as osp
def replace(file, old_content, new_content):
content = read_file(file)
content = content.replace(old_content, new_content)
rewrite_file(file, content)
# 读文件内容
def read_file(file):
with open(file, encoding='UTF-8') as f:
read_all = f.read()
f.close()
return read_all
# 写内容到文件
def rewrite_file(file, data):
with open(file, 'w', encoding='UTF-8') as f:
f.write(data)
f.close()
# src_data='/media/ckq/data/kitti/data_object/training/label_2'
src_data='/media/ckq/NewData/test/MOT/images/train'
seqs = [s for s in os.listdir(src_data)]
seqs.sort() #排个序
print(seqs)
for seq in seqs:
path=osp.join(src_data,seq,'gt/gt.txt')
print(path)
replace(path, ' ', ',')
2.图片重命名 原先是从0开始，重新命名后是img000001—img0000…
import os
import re
import sys
import os
import os.path as osp
def renameall(path):
fileList = os.listdir(path) # 待修改文件夹
fileList.sort()
print("修改前：" + str(fileList)) # 输出文件夹中包含的文件
os.chdir(path) # 将当前工作目录修改为待修改文件夹的位置
num = 1 # 名称变量
for fileName in fileList: # 遍历文件夹中所有文件
#print(fileName)
tiqu_fileName=fileName[:6]
# tiqu_fileName=fileName[3:6]
seq_num =int(tiqu_fileName)+1
# seq_num = int(tiqu_fileName)
#print(seq_num)
name = str(seq_num).zfill(6) # 设置宽度
all_name='img'+name
print(all_name)
# pat = ".+\.(jpg|jpeg|JPG)" # 匹配文件名正则表达式
pat = ".+\.(png|PNG)" # 匹配文件名正则表达式
pattern = re.findall(pat, fileName) # 进行匹配
print('pattern[0]:', pattern)
print('num：', num, 'filename:', fileName)
os.rename(fileName, (all_name + '.' + pattern[0])) # 文件重新命名
#name = str(num).zfill(6) # 设置宽度
#name = num
#os.rename(fileName, ('img' + name +'.'+ pattern[0])) # 文件重新命名
#print(name)
#os.rename(fileName, (name + '.' + pattern[0])) # 文件重新命名
#num = num + 1 # 改变编号，继续下一项
print("---------------------------------------------------")
sys.stdin.flush() # 刷新
print("修改后：" + str(os.listdir(path))) # 输出修改后文件夹中包含的文件
src_path='/media/ckq/NewData/test/MOT/images/train'
imgs_name = os.listdir(src_path)
imgs_name.sort()
print(imgs_name)
for img_name in imgs_name:
img_path=osp.join(src_path,img_name,'img1')
print(img_path)
renameall(img_path)
3.生成训练gt，每张图片对应一个gt.txt
对原先的改动是：首先给每张图片对应的文档建立一个空文件里面预先赋值0,0,0,0,0,0
之后在这基础上对图片添加检测交通灯的信息
import os.path as osp
import os
import shutil
import numpy as np
def mkdirs(d):
# if not osp.exists(d):
if not osp.isdir(d):
os.makedirs(d)
data_root = '/media/ckq/NewData/test/'
seq_root = data_root + 'MOT/images/train'
label_root = data_root + 'MOT/labels_with_ids/train'
if not os.path.isdir(label_root):
mkdirs(label_root)
else: # 如果之前已经生成过: 递归删除目录和文件, 重新生成目录
shutil.rmtree(label_root)
os.makedirs(label_root)
#这次只有交通灯
# cls_map = {
# 'Person=1'
# 'Pedestrian=2'
# 'Car=3'
# 'Person_sitting=4'
# 'Cyclist=5'
# 'Van=6'
# 'Truck=7'
# 'Tram=8'
# 'Misc=9'
# 'DontCare=10'
# }
print("Dir %s made" % label_root)
seqs = [s for s in os.listdir(seq_root)]
#seqs=['0000']
# seqs=['0000', '0001', '0002', '0003',
# '0004', '0005', '0006', '0007',
# '0008', '0009', '0010', '0011',
# '0012', '0014', '0015', '0018',
# '0019', '0020']
#打印序列
print(seqs)
tid_curr = 0
tid_last = -1
total_track_id_num = 0
for seq in seqs: # 每段视频都对应一个gt.txt
print("Process %s, " % seq, end='')
seq_info_path = osp.join(seq_root, seq, 'seqinfo.ini') #提取每个数据的info信息 /media/ckq/data/kitti/MOT/images/train
print(seq_info_path)
with open(seq_info_path) as seq_info_h: # 读取 *.ini 文件
seq_info = seq_info_h.read()
seq_photo_numbers = int(seq_info[seq_info.find('seqLength=') + 10:seq_info.find('\nimWidth')]) # 视频的宽
seq_width = int(seq_info[seq_info.find('imWidth=') + 8:seq_info.find('\nimHeight')]) # 视频的宽
seq_height = int(seq_info[seq_info.find('imHeight=') + 9:seq_info.find('\nimExt')]) # 视频的高
print('seq_photo_numbers:', seq_photo_numbers)
print('seq_width:',seq_width)
print('seq_height:', seq_height)
gt_txt = osp.join(seq_root, seq, 'gt', 'gt.txt') # 读取GT文件
# print(gt_txt) #打印路径
#gt = np.loadtxt(gt_txt, dtype=np.str, delimiter=',') # 加载成np格式
gt = np.loadtxt(gt_txt, dtype=np.float64, delimiter=',') # 加载成np格式
# print(gt) #打印文本内容
# print('gt.T')
# print(gt.T) #也是打印文本内容
idx = np.lexsort(gt.T[:2, :]) # 优先按照track id排序(对视频帧进行排序, 而后对轨迹ID进行排序)
# print(idx)
gt = gt[idx, :]
tr_ids = set(gt[:, 1])
# print("%d track ids in seq %s" % (len(tr_ids), seq))
total_track_id_num += len(tr_ids) # track id统计数量如何正确计算？
seq_label_root = osp.join(label_root, seq, 'img1')
mkdirs(seq_label_root)
for i in range(1,seq_photo_numbers+1):
seq_name = '{:06d}.txt'.format(i)
add_name = 'img' + str(seq_name)
print(add_name)
label_f_path = osp.join(seq_label_root, add_name) # 原先不加1 因为kitti数据集从第一帧从0 开始所以加了1 同时加上img前缀名称
#os.mknod(label_f_path) #创建空文件
file = open(label_f_path, 'a')
label_str = '0 {:d} {:.6f} {:.6f} {:.6f} {:.6f}\n'.format(
0,
0, # center_x
0, # center_y
0, # bbox_w
0) # bbox_h
file.write(label_str)
# 读取GT数据的每一行(一行即一条数据)
# for obj_id det_id kf_id x_b y_b w_b h_b
for obj_id, det_id, kf_id, min_x, min_y, max_x, max_y in gt:
fid = int(kf_id+1)
tid = int(det_id)
#
# 判断是否是同一个track, 记录上一个track和当前track
if not tid == tid_last: # not 的优先级比 == 高
tid_curr += 1
tid_last = tid
#由于kitti标签与训练标签参数有点不同 需要自己计算 x y w h
# w=float(bbox_right-bbox_left)
# h=float(bbox_bottom-bbox_top)
# x=int(bbox_left+0.5)
# y=int(bbox_top+0.5)
w = float(max_x-min_x)
h = float(max_y-min_y)
x = int(min_x+0.5)
y = int(min_y+0.5)
# min_x, min_y, max_x, max_y
# x_b, y_b, w_b, h_b
# bbox中心点坐标
x += w / 2
y += h / 2
# 网label中写入track id, bbox中心点坐标和宽高(归一化到0~1)
# 第一列的0是默认只对一种类别进行多目标检测跟踪(0是类别)
label_str = '0 {:d} {:.6f} {:.6f} {:.6f} {:.6f}\n'.format(
tid_curr,
x / seq_width, # center_x
y / seq_height, # center_y
w / seq_width, # bbox_w
h / seq_height) # bbox_h
# print(label_str.strip())
seq_name= '{:06d}.txt'.format(fid)
add_name= 'img'+str(seq_name)
print(add_name)
label_f_path = osp.join(seq_label_root,add_name) #原先不加1 因为kitti数据集从第一帧从0 开始所以加了1 同时加上img前缀名称
#label_f_path = osp.join(seq_label_root, '{:06d}.txt'.format(fid+1)) #原先不加1 因为kitti数据集从第一帧从0 开始所以加了1
#label_f_path = osp.join(seq_label_root, '{:06d}.txt'.format(fid))
with open(label_f_path, 'a') as f: # 以追加的方式添加每一帧的label
f.write(label_str)
print("Total %d track ids in this dataset" % total_track_id_num)
print('Done')
4.删除有物体信息的gt，对于没有物体信息的就不删除0,0,0,0,0,0
import fileinput
import os
import re
import sys
import os
import os.path as osp
src_path="/media/ckq/NewData/test/MOT/labels_with_ids/train"
fileList = os.listdir(src_path) # 待修改文件夹
fileList.sort()
# print(fileList)
for filename in fileList:
img_path=osp.join(src_path,filename,'img1')
print(img_path)
img_path_gt_list = os.listdir(img_path) # 待修改文件夹
# print(img_path_gt)
for img_path_gt in img_path_gt_list:
img_gt_name = osp.join(img_path, img_path_gt)
print(img_gt_name)
count=0
f = open(img_gt_name,"r")
for line in f.readlines():
count=count+1
if(count > 1):
with open(img_gt_name, 'r') as r:
lines = r.readlines()
with open(img_gt_name, 'w') as w:
for l in lines:
if '0 0 0.000000 0.000000 0.000000 0.000000' not in l:
w.write(l)
5.展示标签是否正确
# -*- coding:utf-8 -*-
import os
import cv2
import os.path as osp
'''
显示跟踪训练数据集标注
'''
root_path ='/media/ckq/NewData/test/MOT'
img_dir = "images/train"
label_dir = "labels_with_ids/train"
imgs = os.listdir(root_path + "/" + img_dir) #遍历图片数据集列表 0000 0001........
imgs.sort()
for i, img in enumerate(imgs): #一个一个遍历
#img_name = img[:-1] #img[:-1] -1代表从右往左 第一个不取
#print(img)
img_name=img #每个图片集名字
print(img_name)
label_path=osp.join(root_path,label_dir,img_name,'img1')
# print(label_path)
label_gts_name=os.listdir(label_path)
label_gts_name.sort()
# print(label_gts_name)
for frame_gt in label_gts_name:
#print(frame_gt)
frame_gt_name=frame_gt[:9]
#print(frame_gt_name)
label_f = open(label_path + "/" +frame_gt_name+".txt", "r") #路劲标签名
#print(label_f)
lines = label_f.readlines()
# print(lines)
# print(root_path + "/" + img_dir + "/" + img+"/img/"+frame_gt_name)
img_data = cv2.imread(root_path + "/" + img_dir + "/" + img+"/img1/"+frame_gt_name+".png") #gt对应的图片序号
# img_data = cv2.imread(root_path + "/" + img_dir + "/" + img + "/img1/" + frame_gt_name + ".png") # gt对应的图片序号
# print(img_data)
H, W, C = img_data.shape
# print(H)
# print(W)
# print(C)
for line in lines:
line_list = line.strip().split()
class_num = int(line_list[0]) # 类别号
obj_ID = int(line_list[1]) # 目标ID
x, y, w, h = line_list[2:] # 中心坐标，宽高（经过原图宽高归一化后）
x = int(float(x) * W)
y = int(float(y) * H)
w = int(float(w) * W)
h = int(float(h) * H)
left = int(x - w / 2)
top = int(y - h / 2)
right = left + w
bottom = top + h
cv2.circle(img_data, (x, y), 1, (0, 0, 255))
cv2.rectangle(img_data, (left, top), (right, bottom), (0, 255, 0), 2)
cv2.putText(img_data,str(obj_ID), (left, top), cv2.FONT_HERSHEY_COMPLEX, 0.5, (0, 0, 255), 1)
# resized_img = cv2.resize(img_data, (800, 416))
resized_img = cv2.resize(img_data, (1242, 375))
cv2.imshow("label", resized_img)
cv2.waitKey(100)
6.生成train文件
import os
import os.path as osp
image_flder = "/media/ckq/NewData/test/MOT/images/train"
#image_flder = "/home/ckq/Desktop/MOT/images/train" #用来test
need_image_flder="MOT/images/train"
imgs = os.listdir(image_flder)
#print(imgs)
train_f = open("/media/ckq/NewData/test/MOT/unreal_light.train", "w")
#train_f = open("/home/ckq/Desktop/MOT/kitt_car.train", "w") #用来test
for img_name in imgs:
image_path=osp.join(image_flder,img_name,'img1')
need_image_path=osp.join(need_image_flder,img_name,'img1')
#print(image_path)
print(need_image_path)
image_names=os.listdir(image_path)
image_names.sort()
print(image_names)
for image_name in image_names:
#save_str = image_path + '/' + image_name +"\n" #这里保存的路径是绝对路径 代码中不需要绝对路径 相对路径即可
save_str = need_image_path + '/' + image_name + "\n"
print(save_str)
train_f.write(save_str)
train_f.close()