yolov2训练ICDAR2011数据集
首先下载数据集train-textloc.zip
其groundtruth文件如下所示:
158,128,412,182,"Footpath"
442,128,501,170,"To"
393,198,488,240,"and"
63,200,363,242,"Colchester"
71,271,383,313,"Greenstead"
ground truth 文件格式为:xmin, ymin, xmax, ymax, label。同时,要注意,这里的坐标系是如下摆放:
将此txt文件转换成voc xml文件的代码:
icdar2voc.py
#! /usr/bin/python
#-*-coding:utf8-*- import os, sys
import glob
from PIL import Image # ICDAR 图像存储位置
src_img_dir = "train-textloc"
# ICDAR 图像的 ground truth 的 txt 文件存放位置
src_txt_dir = "train-textloc" img_Lists = glob.glob(src_img_dir + '/*.jpg') img_basenames = [] # e.g. 100.jpg
for item in img_Lists:
img_basenames.append(os.path.basename(item)) img_names = [] # e.g. 100
for item in img_basenames:
temp1, temp2 = os.path.splitext(item)
img_names.append(temp1) for img in img_names:
im = Image.open((src_img_dir + '/' + img + '.jpg'))
width, height = im.size # open the crospronding txt file
gt = open(src_txt_dir + '/gt_' + img + '.txt').read().splitlines() # write in xml file
#os.mknod(src_txt_dir + '/' + img + '.xml')
xml_file = open((src_txt_dir + '/' + img + '.xml'), 'w')
xml_file.write('<annotation>\n')
xml_file.write(' <folder>VOC2007</folder>\n')
xml_file.write(' <filename>' + str(img) + '.jpg' + '</filename>\n')
xml_file.write(' <size>\n')
xml_file.write(' <width>' + str(width) + '</width>\n')
xml_file.write(' <height>' + str(height) + '</height>\n')
xml_file.write(' <depth>3</depth>\n')
xml_file.write(' </size>\n') # write the region of text on xml file
for img_each_label in gt:
spt = img_each_label.split(',')
xml_file.write(' <object>\n')
xml_file.write(' <name>text</name>\n')
xml_file.write(' <pose>Unspecified</pose>\n')
xml_file.write(' <truncated>0</truncated>\n')
xml_file.write(' <difficult>0</difficult>\n')
xml_file.write(' <bndbox>\n')
xml_file.write(' <xmin>' + str(spt[0]) + '</xmin>\n')
xml_file.write(' <ymin>' + str(spt[1]) + '</ymin>\n')
xml_file.write(' <xmax>' + str(spt[2]) + '</xmax>\n')
xml_file.write(' <ymax>' + str(spt[3]) + '</ymax>\n')
xml_file.write(' </bndbox>\n')
xml_file.write(' </object>\n') xml_file.write('</annotation>')
再将xml文件转换成yolo的txt格式:
voc_label.py
import xml.etree.ElementTree as ET
import pickle
import os
from os import listdir, getcwd
from os.path import join classes = ["text"] def convert(size, box):
dw = 1./size[0]
dh = 1./size[1]
x = (box[0] + box[1])/2.0
y = (box[2] + box[3])/2.0
w = box[1] - box[0]
h = box[3] - box[2]
x = x*dw
w = w*dw
y = y*dh
h = h*dh
return (x,y,w,h) for i in range(100,329):
in_file = open('train-textloc/%d.xml'% i )
out_file = open('train-textloc/%d.txt'% i , 'w')
tree=ET.parse(in_file)
root = tree.getroot()
size = root.find('size')
w = int(size.find('width').text)
h = int(size.find('height').text) for obj in root.iter('object'):
difficult = obj.find('difficult').text
cls = obj.find('name').text
if cls not in classes or int(difficult) == 1:
continue
cls_id = classes.index(cls)
xmlbox = obj.find('bndbox')
b = (float(xmlbox.find('xmin').text), float(xmlbox.find('xmax').text), float(xmlbox.find('ymin').text), float(xmlbox.find('ymax').text))
bb = convert((w,h), b)
out_file.write(str(cls_id) + " " + " ".join([str(a) for a in bb]) + '\n')
下面开始修改yolo的配置:
把20类改成1类
cfg/voc.data文件中:
- classes 改成1。
- names=data/voc.names。
- voc.names里只写一行 text 即可。
cfg/yolo_voc.cfg文件中 :
- 【region】层中 classes 改成1。
- 【region】层上方第一个【convolution】层,其中的filters值要进行修改,改成(classes+ coords+ 1)* (NUM) ,我的情况中:(1+4+1)* 5=30,我把filters 的值改成了30。
- 修改filters的建议来源自(https://groups.google.com/forum/#!topic/darknet/B4rSpOo84yg),我修改了之后一切正常。
src/yolo.c 文件中 :(经指正,步骤3,4是yolo1的内容,使用yolo_v2的话可以不用更改)
- 位置大约第14行左右改成:char *voc_names={“text”},原来里面有20类的名字,我改成了唯一1类的名字。
- 位置大约第328行左右,修改draw_detection这个函数最后一个参数:20改成1。这个函数用于把系统检测出的框给画出来,并把画完框的图片传回第一个参数im中,用于保存和显示。
- 位置大约第361行左右,demo函数中,倒数第三个参数我把20改成了1,虽然不知道有没有用,反正对结果没什么影响。
src/yolo_kernels.cu 文件中 :(经指正,步骤3,4是yolo1的内容,使用yolo_v2的话可以不用更改)
- 位置第62行,draw_detection这个函数最后一个参数20改成1。
scripts/voc_label.py 文件中(这个应该没用的) :
- 位置第9行改成:classes=[“text”],因为我只有一类。
建立一个文件夹,里面JPEGImages里放入所有的图片,labels里放入所有的标签,系统会自动识别。
然后生成train.txt
list.py
# -*- coding: utf-8 -*-
import os
fw = open('train.txt','w')
files = os.listdir('/home/mingyu_ding/darknet/voc/Table/JPEGImages')
for f in files:
file = '/home/mingyu_ding/darknet/voc/Table/JPEGImages' + os.sep + f
print >> fw, file
就可以开始训练了,系统默认会迭代45000次。
nohup ./darknet detector train cfg/voc.data cfg/yolo-voc.cfg darknet19_448.conv.23 > log.txt &
当然迭代次数是可以修改的,应该是在cfg/yolo_voc.cfg修改max_batches的值就行。
没训练完就可以测试啦
./darknet detector test cfg/voc.data cfg/yolo-voc.cfg backup/yolo-voc_5000.weights ../Downloads/1.jpg
参考链接:http://blog.csdn.net/hysteric314/article/details/54097845
结果如下:
后来又训练了京东的参数图数据。
图是自己下载的,重命名后用labelimg进行标注,之后用voc_label.py修改成标签数据即可。
rename.py
import os
from os.path import join files = os.listdir('imageset')
i = 0
for f in files:
i += 1
print os.path.join(os.getcwd() + os.sep + 'imageset' , f)
print os.getcwd() + os.sep + 'imageset' + os.sep + '%d.jpg' % i
os.rename(os.path.join(os.getcwd() + os.sep + 'imageset' , f), os.getcwd() + os.sep + 'imageset' + os.sep + '%d.jpg' % i)
训练办法和上面一模一样'
结果如下图:
想要输出预测框的位置的话
修改 image.c 后重新 make 就可以了
printf("left:%d right:%d top:%d bot:%d\n",left,right,top,bot);
修改 detector.c 后 make 可以切割出想要的位置,如下图所示。
在函数draw_detections() 前面修改就可以
int i;
int j = ;
for(i = ; i < l.w*l.h*l.n; ++i){
int class = max_index(probs[i], l.classes);
float prob = probs[i][class];
if(prob > thresh){
box b = boxes[i];
int left = (b.x-b.w/.)*im.w;
int right = (b.x+b.w/.)*im.w;
int top = (b.y-b.h/.)*im.h;
int bot = (b.y+b.h/.)*im.h;
if(left < ) left = ;
if(right > im.w-) right = im.w-;
if(top < ) top = ;
if(bot > im.h-) bot = im.h-;
int width = right - left;
int height = bot - top;
IplImage* src = cvLoadImage(input,-);
CvSize size = cvSize(width, height);
//printf("%d,%d",src->depth,src->nChannels);
IplImage* roi = cvCreateImage(size,src->depth,src->nChannels);
CvRect box = cvRect(left, top, size.width, size.height);
cvSetImageROI(src,box);
cvCopy(src,roi,NULL);
//cvNamedWindow("pic",CV_WINDOW_AUTOSIZE);
//cvShowImage("pic",src);
char name[] = "cut";
char name1[] = ".jpg";
char newname[];
sprintf(newname,"%s%d_%.0f%s",name,j,*prob,name1);
//printf("%s\n",newname);
j++;
cvSaveImage(newname,roi,);
//cvWaitKey(0);
//cvDestoryWindow("pic");
cvReleaseImage(&src);
cvReleaseImage(&roi);
//printf("left:%d right:%d top:%d bot:%d\n",left,right,top,bot);
}
};
aaarticlea/png;base64,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" alt="" />
下一步就是识别出参数框里的文字了,需要数据集和标签的可以联系我。
sudo apt-get install tesseract-ocr
sudo apt-get install tesseract-ocr-chi-sim
tesseract cut1_93.jpg out -l eng+chi_sim
out.txt 就可以看了
yolov2训练ICDAR2011数据集的更多相关文章
- Fast RCNN 训练自己数据集 (1编译配置)
FastRCNN 训练自己数据集 (1编译配置) 转载请注明出处,楼燚(yì)航的blog,http://www.cnblogs.com/louyihang-loves-baiyan/ https:/ ...
- 使用caffe训练mnist数据集 - caffe教程实战(一)
个人认为学习一个陌生的框架,最好从例子开始,所以我们也从一个例子开始. 学习本教程之前,你需要首先对卷积神经网络算法原理有些了解,而且安装好了caffe 卷积神经网络原理参考:http://cs231 ...
- 实践详细篇-Windows下使用VS2015编译的Caffe训练mnist数据集
上一篇记录的是学习caffe前的环境准备以及如何创建好自己需要的caffe版本.这一篇记录的是如何使用编译好的caffe做训练mnist数据集,步骤编号延用上一篇 <实践详细篇-Windows下 ...
- 使用py-faster-rcnn训练VOC2007数据集时遇到问题
使用py-faster-rcnn训练VOC2007数据集时遇到如下问题: 1. KeyError: 'chair' File "/home/sai/py-faster-rcnn/tools/ ...
- (转)理解YOLOv2训练过程中输出参数含义
最近有人问起在YOLOv2训练过程中输出在终端的不同的参数分别代表什么含义,如何去理解这些参数?本篇文章中我将尝试着去回答这个有趣的问题. 刚好现在我正在训练一个YOLOv2模型,拿这个真实的例子来讨 ...
- 理解YOLOv2训练过程中输出参数含义
原英文地址: https://timebutt.github.io/static/understanding-yolov2-training-output/ 最近有人问起在YOLOv2训练过程中输出在 ...
- YOLOV4在linux下训练自己数据集(亲测成功)
最近推出了yolo-v4我也准备试着跑跑实验看看效果,看看大神的最新操作 这里不做打标签工作和配置cuda工作,需要的可以分别百度搜索 VOC格式数据集制作,cuda和cudnn配置 我们直接利用 ...
- Scaled-YOLOv4 快速开始,训练自定义数据集
代码: https://github.com/ikuokuo/start-scaled-yolov4 Scaled-YOLOv4 代码: https://github.com/WongKinYiu/S ...
- ubuntu yolov2 训练自己的数据集
项目需求+锻炼自己,尝试用yolov2跑自己的数据集,中间遇到了很多问题,记下来防止忘记 一.数据集 首先发现由于物体特殊没有合适的现成的数据集使用,所以只好自己标注,为了减少工作量,先用opencv ...
随机推荐
- 蓝桥网试题 java 入门训练 A+B问题
---------------------------------------------------------------------------------------------------- ...
- 源码(07) -- java.util.Iterator<E>
java.util.Iterator<E> 源码分析(JDK1.7) ----------------------------------------------------------- ...
- java udp 发送小数数字(较难)
代码全部来自:http://825635381.iteye.com/blog/2046882,在这里非常感谢了,我运行测试了下,非常正确,谢谢啊 服务端程序: package udpServer; i ...
- HTTP协议详解【转】
当今web程序的开发技术真是百家争鸣,ASP.NET, PHP, JSP,Perl, AJAX 等等. 无论Web技术在未来如何发展,理解Web程序之间通信的基本协议相当重要, 因为它让我们理解了We ...
- TCP协议详解
TCP协议详解 一.TCP协议 1.TCP 通过以下方式提供可靠性: · ◆ 应用程序分割为TCP认为最合适发送的数据块.由TCP传递给IP的信息单位叫做报文段. · ◆ 当TCP发出一个报文段后 ...
- API模板
#include <windows.h> #include <windowsx.h> #define DIVISIONS 5 LRESULT CALLBACK WndProc ...
- [APUE]进程关系(上)
一.终端登录 1. 4.3+BSD终端登录 系统管理员创建一个通常名为/etc/ttys的文件,其中,每个终端设备有一行,每一行说明设备名和传到getty程序的参数,这些参数说明了终端的波特率.当系统 ...
- Python从入门到放弃之迭代器
迭代器是Python2.1中新加入的接口(PEP 234),说明如下: The iterator provides a 'get next value' operation that produces ...
- ubuntu14.04 + OpenCV2.4.9 配置方法
1. 安装openCV 所需依赖库或软件: sudo apt-get install build-essential cmake libgtk2.0-dev pkg-config python-de ...
- BeautifulSoup简述
网页解析器 从网页中提取有价值数据的工具 网页解析器种类 正则表达式 (模糊匹配) html.parser (结构化解析) BeautifulSoup第三方插件 (结构化解析,相对比较强大) lxml ...