CCPD数据集

2024-06-17 04:44
文章标签 数据 ccpd

本文主要是介绍CCPD数据集，希望对大家解决编程问题提供一定的参考价值，需要的开发者们随着小编来一起学习吧！

官网：https://github.com/detectRecog/CCPD

其它介绍：https://blog.csdn.net/qianbin3200896/article/details/103009221

CCPD (Chinese City Parking Dataset, ECCV)

provinces = ["皖", "沪", "津", "渝", "冀", "晋", "蒙", "辽", "吉", "黑", "苏", "浙", "京", "闽", "赣", "鲁", "豫", "鄂", "湘", "粤", "桂", "琼", "川", "贵", "云", "藏", "陕", "甘", "青", "宁", "新", "警", "学", "O"]
alphabets = ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'J', 'K', 'L', 'M', 'N', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W','X', 'Y', 'Z', 'O']
ads = ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'J', 'K', 'L', 'M', 'N', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X','Y', 'Z', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 'O']

官网github给的3个列表最后都有一个字母O，不是数字0，当作结尾符号，因为中国车牌没有大写字母O，其实也没有大写字母I

省份简称里的"警" 公安内部车， "学" 驾校的车，车牌号没有小写字母

图片名称例子

name = "025-95_113-154&383_386&473-386&473_177&454_154&383_363&402-0_0_22_27_27_33_16-37-15.jpg"print(len(name.split('-')))
print(name.split('-'))

结果如下

7
['025', '95_113', '154&383_386&473', '386&473_177&454_154&383_363&402', '0_0_22_27_27_33_16', '37', '15.jpg']

分割成了7段

  • 154&383_386&473：车牌框的第1、第4个点
  • 0_0_22_27_27_33_16：省份简称_字母_5个字母或数字组合

2020年绿牌数据

车牌号有8位 

train数量

5769
['./CCPD2020/ccpd_green/train/0245182291667-88_93-221&481_490&573-489&561_221&573_223&485_490&481-0_0_3_25_29_31_31_31-132-139.jpg','./CCPD2020/ccpd_green/train/0161783854167-93_97-255&489_465&566-465&566_258&553_255&489_461&495-0_0_3_24_27_26_31_30-148-296.jpg','./CCPD2020/ccpd_green/train/0133203125-90_102-228&515_426&583-426&583_240&582_228&518_413&515-0_0_5_24_29_33_33_30-121-38.jpg']
val数量
import globval_dir = './CCPD2020/ccpd_green/val'
val_imgPaths = glob.glob(val_dir+'/*.jpg')
print(len(val_imgPaths))
print(val_imgPaths[0])
1001
./CCPD2020/ccpd_green/val/04189453125-105_107-165&464_435&620-435&620_172&540_165&464_433&532-0_0_3_25_29_29_30_30-95-92.jpg
test数量
import globtest_dir = './CCPD2020/ccpd_green/test'
test_imgPaths = glob.glob(test_dir+'/*.jpg')
print(len(test_imgPaths))
print(test_imgPaths[0])
5006
./CCPD2020/ccpd_green/test/00954022988505747-90_263-190&522_356&574-356&574_195&571_190&522_351&523-0_0_3_29_32_26_26_26-151-54.jpg
画框

用上面这张图为例

import osimgPath = 'CCPD2020/ccpd_green/val/02-84_90-245&479_437&584-437&551_245&584_245&503_429&479-0_0_3_24_33_26_33_26-103-32.jpg'print(os.path.splitext(imgPath))

打印如下

('CCPD2020/ccpd_green/val/02-84_90-245&479_437&584-437&551_245&584_245&503_429&479-0_0_3_24_33_26_33_26-103-32', '.jpg')

分割图片名称

import osimgPath = 'CCPD2020/ccpd_green/val/02-84_90-245&479_437&584-437&551_245&584_245&503_429&479-0_0_3_24_33_26_33_26-103-32.jpg'numName, _ = os.path.splitext(imgPath)
print(numName.split('-'))

打印如下

['CCPD2020/ccpd_green/val/02', '84_90', '245&479_437&584', '437&551_245&584_245&503_429&479', '0_0_3_24_33_26_33_26', '103', '32']

分割框的2个坐标

import osimgPath = 'CCPD2020/ccpd_green/val/02-84_90-245&479_437&584-437&551_245&584_245&503_429&479-0_0_3_24_33_26_33_26-103-32.jpg'numName, _ = os.path.splitext(imgPath)
x1_y1,x2_y2 = numName.split('-')[2].split('_')
print(x1_y1,x2_y2)
x1,y1 = x1_y1.split('&')
x2,y2 = x2_y2.split('&')
print(x1,y1,x2,y2)

打印如下

245&479 437&584
245 479 437 584

显示框

import os
import cv2imgPath = 'CCPD2020/ccpd_green/val/02-84_90-245&479_437&584-437&551_245&584_245&503_429&479-0_0_3_24_33_26_33_26-103-32.jpg'numName, _ = os.path.splitext(imgPath)
x1_y1,x2_y2 = numName.split('-')[2].split('_')
x1,y1 = x1_y1.split('&')
x2,y2 = x2_y2.split('&')img = cv2.imread(imgPath)
cv2.rectangle(img, (int(x1),int(y1)), (int(x2), int(y2)), (0, 0, 255), 2)
cv2.imshow('img',img)
cv2.waitKey(10000)
cv2.destroyAllWindows()

车牌号名称
import os
import cv2imgPath = 'CCPD2020/ccpd_green/val/02-84_90-245&479_437&584-437&551_245&584_245&503_429&479-0_0_3_24_33_26_33_26-103-32.jpg'numName, _ = os.path.splitext(imgPath)
print(numName.split('-')[4])
print(numName.split('-')[4].split('_'))

打印如下

0_0_3_24_33_26_33_26
['0', '0', '3', '24', '33', '26', '33', '26']

换成号码

import os
import cv2provinces = ["皖", "沪", "津", "渝", "冀", "晋", "蒙", "辽", "吉", "黑", "苏", "浙", "京", "闽", "赣", "鲁", "豫", "鄂", "湘", "粤", "桂", "琼", "川", "贵", "云", "藏", "陕", "甘", "青", "宁", "新", "警", "学", "O"]
alphabets = ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'J', 'K', 'L', 'M', 'N', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W','X', 'Y', 'Z', 'O']
ads = ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'J', 'K', 'L', 'M', 'N', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X','Y', 'Z', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 'O']imgPath = 'CCPD2020/ccpd_green/val/02-84_90-245&479_437&584-437&551_245&584_245&503_429&479-0_0_3_24_33_26_33_26-103-32.jpg'numName, _ = os.path.splitext(imgPath)
index = [int(i) for i in numName.split('-')[4].split('_')]
first_index = index[0]
second_index = index[1]
last5_index = index[2:]print(provinces[first_index],alphabets[second_index])
print([ads[i] for i in last5_index])

打印如下

皖 A
['D', '0', '9', '2', '9', '2']

拼成号码
import os
import cv2provinces = ["皖", "沪", "津", "渝", "冀", "晋", "蒙", "辽", "吉", "黑", "苏", "浙", "京", "闽", "赣", "鲁", "豫", "鄂", "湘", "粤", "桂", "琼", "川", "贵", "云", "藏", "陕", "甘", "青", "宁", "新", "警", "学", "O"]
alphabets = ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'J', 'K', 'L', 'M', 'N', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W','X', 'Y', 'Z', 'O']
ads = ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'J', 'K', 'L', 'M', 'N', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X','Y', 'Z', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 'O']imgPath = 'CCPD2020/ccpd_green/val/02-84_90-245&479_437&584-437&551_245&584_245&503_429&479-0_0_3_24_33_26_33_26-103-32.jpg'numName, _ = os.path.splitext(imgPath)
index = [int(i) for i in numName.split('-')[4].split('_')]
first_index = index[0]
second_index = index[1]
last5_index = index[2:]s = ''
for i in [ads[i] for i in last5_index]:s += iprint(provinces[first_index]+alphabets[second_index]+s)

打印如下

皖AD09292

裁剪车牌保存
import os
import cv2provinces = ["皖", "沪", "津", "渝", "冀", "晋", "蒙", "辽", "吉", "黑", "苏", "浙", "京", "闽", "赣", "鲁", "豫", "鄂", "湘", "粤", "桂", "琼", "川", "贵", "云", "藏", "陕", "甘", "青", "宁", "新", "警", "学", "O"]
alphabets = ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'J', 'K', 'L', 'M', 'N', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W','X', 'Y', 'Z', 'O']
ads = ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'J', 'K', 'L', 'M', 'N', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X','Y', 'Z', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 'O']imgPath = 'CCPD2020/ccpd_green/val/02-84_90-245&479_437&584-437&551_245&584_245&503_429&479-0_0_3_24_33_26_33_26-103-32.jpg'numName, _ = os.path.splitext(imgPath)
index = [int(i) for i in numName.split('-')[4].split('_')]
first_index = index[0]
second_index = index[1]
last5_index = index[2:]
s = ''
for i in [ads[i] for i in last5_index]: s += i
imgName = provinces[first_index]+alphabets[second_index]+s+'.jpg'x1_y1,x2_y2 = numName.split('-')[2].split('_')
x1,y1 = x1_y1.split('&')
x2,y2 = x2_y2.split('&')
img = cv2.imread(imgPath)
img_crop = img[int(y1):int(y2),int(x1):int(x2)]
cv2.imwrite(imgName,img_crop)

train/val/test的图片是否重名
import globtrain_dir = './CCPD2020/ccpd_green/train'
train_imgPaths = glob.glob(train_dir+'/*.jpg')val_dir = './CCPD2020/ccpd_green/val'
val_imgPaths = glob.glob(val_dir+'/*.jpg')test_dir = './CCPD2020/ccpd_green/test'
test_imgPaths = glob.glob(test_dir+'/*.jpg')print(len(train_imgPaths)+len(val_imgPaths)+len(test_imgPaths))merge = set()
merge.update(train_imgPaths,val_imgPaths,test_imgPaths)
print(len(merge))

打印如下

11776
11776

没有重名的图片

合并train/val/test文件夹

把train/val/test文件夹下的图片拷到green文件夹

11776
 同一车牌是否多次采集
import glob
import os
import cv2provinces = ["皖", "沪", "津", "渝", "冀", "晋", "蒙", "辽", "吉", "黑", "苏", "浙", "京", "闽", "赣", "鲁", "豫", "鄂", "湘", "粤", "桂", "琼", "川", "贵", "云", "藏", "陕", "甘", "青", "宁", "新", "警", "学", "O"]
alphabets = ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'J', 'K', 'L', 'M', 'N', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W','X', 'Y', 'Z', 'O']
ads = ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'J', 'K', 'L', 'M', 'N', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X','Y', 'Z', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 'O']def get_chepai(path):numName, _ = os.path.splitext(path)index = [int(i) for i in numName.split('-')[4].split('_')]first_index = index[0]second_index = index[1]last5_index = index[2:]s = ''for i in [ads[i] for i in last5_index]: s += ireturn provinces[first_index]+alphabets[second_index]+sgreen_dir = './CCPD2020/ccpd_green/green'
green_imgPaths = glob.glob(green_dir+'/*.jpg')
print('图片数量:',len(green_imgPaths))chepai = set()
for imgPath in green_imgPaths:chepai.add(get_chepai(imgPath))
print('车牌数量:', len(chepai))

打印如下

图片数量: 11776
车牌数量: 3298

同一车牌采集了多次

批量裁剪车牌
import glob
import os
import cv2provinces = ["皖", "沪", "津", "渝", "冀", "晋", "蒙", "辽", "吉", "黑", "苏", "浙", "京", "闽", "赣", "鲁", "豫", "鄂", "湘", "粤", "桂", "琼", "川", "贵", "云", "藏", "陕", "甘", "青", "宁", "新", "警", "学", "O"]
alphabets = ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'J', 'K', 'L', 'M', 'N', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W','X', 'Y', 'Z', 'O']
ads = ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'J', 'K', 'L', 'M', 'N', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X','Y', 'Z', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 'O']def get_chepai(path):numName, _ = os.path.splitext(path)index = [int(i) for i in numName.split('-')[4].split('_')]first_index = index[0]second_index = index[1]last5_index = index[2:]s = ''for i in [ads[i] for i in last5_index]: s += ireturn provinces[first_index]+alphabets[second_index]+sdef get_box(path):numName, _ = os.path.splitext(path)x1_y1,x2_y2 = numName.split('-')[2].split('_')x1,y1 = x1_y1.split('&')x2,y2 = x2_y2.split('&')return [int(x1),int(y1),int(x2),int(y2)]green_dir = './CCPD2020/ccpd_green/green'
green_imgPaths = glob.glob(green_dir+'/*.jpg')
print('图片数量:',len(green_imgPaths))name2num = dict()
for imgPath in green_imgPaths:chepai = get_chepai(imgPath)if chepai in name2num: name2num[chepai] += 1else: name2num[chepai] = 0crop_dir = './CCPD2020/ccpd_green/green_crop/'
for index,imgPath in enumerate(green_imgPaths):chepai = get_chepai(imgPath)img = cv2.imread(imgPath)x1,y1,x2,y2 = get_box(imgPath)img_crop = img[y1:y2,x1:x2]cv2.imwrite(crop_dir+'{}_{}.jpg'.format(chepai,name2num[chepai]),img_crop)name2num[chepai] -= 1if index % 2000==0: print(index)
图片数量: 11776
0
2000
4000
6000
8000
10000
import globcrop_dir = './CCPD2020/ccpd_green/green_crop'
crop_imgPaths = glob.glob(crop_dir+'/*.jpg')
print('裁剪的车牌图片数量:',len(crop_imgPaths))
裁剪的车牌图片数量: 11776

批量裁剪成94x24大小
import glob
import os
import cv2provinces = ["皖", "沪", "津", "渝", "冀", "晋", "蒙", "辽", "吉", "黑", "苏", "浙", "京", "闽", "赣", "鲁", "豫", "鄂", "湘", "粤", "桂", "琼", "川", "贵", "云", "藏", "陕", "甘", "青", "宁", "新", "警", "学", "O"]
alphabets = ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'J', 'K', 'L', 'M', 'N', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W','X', 'Y', 'Z', 'O']
ads = ['A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', 'J', 'K', 'L', 'M', 'N', 'P', 'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X','Y', 'Z', '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 'O']def get_chepai(path):numName, _ = os.path.splitext(path)index = [int(i) for i in numName.split('-')[4].split('_')]first_index = index[0]second_index = index[1]last5_index = index[2:]s = ''for i in [ads[i] for i in last5_index]: s += ireturn provinces[first_index]+alphabets[second_index]+sdef get_box(path):numName, _ = os.path.splitext(path)x1_y1,x2_y2 = numName.split('-')[2].split('_')x1,y1 = x1_y1.split('&')x2,y2 = x2_y2.split('&')return [int(x1),int(y1),int(x2),int(y2)]green_dir = './CCPD2020/ccpd_green/green'
green_imgPaths = glob.glob(green_dir+'/*.jpg')
print('图片数量:',len(green_imgPaths))name2num = dict()
for imgPath in green_imgPaths:chepai = get_chepai(imgPath)if chepai in name2num: name2num[chepai] += 1else: name2num[chepai] = 0crop_dir = './CCPD2020/ccpd_green/green_crop_94x24/'
for index,imgPath in enumerate(green_imgPaths):chepai = get_chepai(imgPath)img = cv2.imread(imgPath)x1,y1,x2,y2 = get_box(imgPath)img_crop = img[y1:y2,x1:x2]img_crop = cv2.resize(img_crop,(94,24))cv2.imwrite(crop_dir+'{}_{}.jpg'.format(chepai,name2num[chepai]),img_crop)name2num[chepai] -= 1if index % 2000==0: print(index)crop_imgPaths = glob.glob(crop_dir+'/*.jpg')
print('裁剪的车牌图片数量:',len(crop_imgPaths))

打印结果

图片数量: 11776
0
2000
4000
6000
8000
10000
裁剪的车牌图片数量: 11776

这篇关于CCPD数据集的文章就介绍到这儿，希望我们推荐的文章对编程师们有所帮助！


http://www.chinasem.cn/article/1068507。

相关文章

MyBatis-Plus通用中等、大量数据分批查询和处理方法

MyBatis-Plus通用中等、大量数据分批查询和处理方法

《MyBatis-Plus通用中等、大量数据分批查询和处理方法》文章介绍MyBatis-Plus分页查询处理,通过函数式接口与Lambda表达式实现通用逻辑,方法抽象但功能强大,建议扩展分批处理及流式... 目录函数式接口获取分页数据接口数据处理接口通用逻辑工具类使用方法简单查询自定义查询方法总结函数式接口
阅读更多...
SQL中如何添加数据(常见方法及示例)

SQL中如何添加数据(常见方法及示例)

《SQL中如何添加数据(常见方法及示例)》SQL全称为StructuredQueryLanguage,是一种用于管理关系数据库的标准编程语言,下面给大家介绍SQL中如何添加数据,感兴趣的朋友一起看看吧... 目录在mysql中，有多种方法可以添加数据。以下是一些常见的方法及其示例。1. 使用INSERT I
阅读更多...
Python使用vllm处理多模态数据的预处理技巧

Python使用vllm处理多模态数据的预处理技巧

《Python使用vllm处理多模态数据的预处理技巧》本文深入探讨了在Python环境下使用vLLM处理多模态数据的预处理技巧,我们将从基础概念出发,详细讲解文本、图像、音频等多模态数据的预处理方法,... 目录1. 背景介绍1.1 目的和范围1.2 预期读者1.3 文档结构概述1.4 术语表1.4.1 核
阅读更多...
MySQL 删除数据详解(最新整理)

MySQL 删除数据详解(最新整理)

《MySQL删除数据详解(最新整理)》：本文主要介绍MySQL删除数据的相关知识,本文通过实例代码给大家介绍的非常详细,对大家的学习或工作具有一定的参考借鉴价值,需要的朋友参考下吧... 目录一、前言二、mysql 中的三种删除方式1.DELETE语句✅ 基本语法： 示例：2.TRUNCATE语句✅ 基本语
阅读更多...
MyBatisPlus如何优化千万级数据的CRUD

MyBatisPlus如何优化千万级数据的CRUD

《MyBatisPlus如何优化千万级数据的CRUD》最近负责的一个项目,数据库表量级破千万,每次执行CRUD都像走钢丝,稍有不慎就引起数据库报警,本文就结合这个项目的实战经验,聊聊MyBatisPl... 目录背景一、MyBATis Plus 简介二、千万级数据的挑战三、优化 CRUD 的关键策略1. 查
阅读更多...
python实现对数据公钥加密与私钥解密

python实现对数据公钥加密与私钥解密

《python实现对数据公钥加密与私钥解密》这篇文章主要为大家详细介绍了如何使用python实现对数据公钥加密与私钥解密,文中的示例代码讲解详细,感兴趣的小伙伴可以跟随小编一起学习一下... 目录公钥私钥的生成使用公钥加密使用私钥解密公钥私钥的生成这一部分，使用python生成公钥与私钥，然后保存在两个文
阅读更多...
mysql中的数据目录用法及说明

mysql中的数据目录用法及说明

《mysql中的数据目录用法及说明》：本文主要介绍mysql中的数据目录用法及说明,具有很好的参考价值,希望对大家有所帮助,如有错误或未考虑完全的地方,望不吝赐教... 目录1、背景2、版本3、数据目录4、总结1、背景安装mysql之后，在安装目录下会有一个data目录，我们创建的数据库、创建的表、插入的
阅读更多...
Navicat数据表的数据添加,删除及使用sql完成数据的添加过程

Navicat数据表的数据添加,删除及使用sql完成数据的添加过程

《Navicat数据表的数据添加,删除及使用sql完成数据的添加过程》：本文主要介绍Navicat数据表的数据添加,删除及使用sql完成数据的添加过程,具有很好的参考价值,希望对大家有所帮助,如有... 目录Navicat数据表数据添加,删除及使用sql完成数据添加选中操作的表则出现如下界面，查看左下角从左
阅读更多...
SpringBoot中4种数据水平分片策略

SpringBoot中4种数据水平分片策略

《SpringBoot中4种数据水平分片策略》数据水平分片作为一种水平扩展策略,通过将数据分散到多个物理节点上,有效解决了存储容量和性能瓶颈问题,下面小编就来和大家分享4种数据分片策略吧... 目录一、前言二、哈希分片2.1 原理2.2 SpringBoot实现2.3 优缺点分析2.4 适用场景三、范围分片
阅读更多...
Redis分片集群、数据读写规则问题小结

Redis分片集群、数据读写规则问题小结

《Redis分片集群、数据读写规则问题小结》本文介绍了Redis分片集群的原理,通过数据分片和哈希槽机制解决单机内存限制与写瓶颈问题,实现分布式存储和高并发处理,但存在通信开销大、维护复杂及对事务支持... 目录一、分片集群解android决的问题二、分片集群图解 分片集群特征如何解决的上述问题？（与哨兵模
阅读更多...

Copyright China编程 23002807@qq.com

沪ICP备2023033072号-2