cc92yy3344 commited on
Commit
cb9e714
·
1 Parent(s): c30ac13

更新说明

Browse files
Files changed (1) hide show
  1. README.md +107 -1
README.md CHANGED
@@ -1,3 +1,109 @@
1
  ---
2
- license: apache-2.0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ annotations_creators:
3
+ - crowdsourced
4
+ language:
5
+ - zh
6
+ language_creators:
7
+ - found
8
+ license:
9
+ - apache-2.0
10
+ multilinguality:
11
+ - monolingual
12
+ pretty_name: "15\u79CD\u852C\u83DC\u6570\u636E\u96C6"
13
+ size_categories:
14
+ - 10K<n<100K
15
+ source_datasets:
16
+ - original
17
+ tags:
18
+ - "\u852C\u83DC"
19
+ - "\u56FE\u50CF\u5206\u7C7B"
20
+ task_categories:
21
+ - image-classification
22
+ task_ids:
23
+ - multi-class-image-classification
24
+ dataset_info:
25
+ features:
26
+ - name: image
27
+ dtype: image
28
+ - name: category
29
+ dtype: int64
30
  ---
31
+
32
+ ## 蔬菜图像数据集
33
+ ### 背景
34
+ 最初的实验是用世界各地发现的15种常见蔬菜进行的。实验选择的蔬菜有:豆类、苦瓜、葫芦、茄子、西兰花、卷心菜、辣椒、胡萝卜、花椰菜、黄瓜、木瓜、土豆、南瓜、萝卜和番茄。共使用了来自15个类的21000张图像,其中每个类包含1400张尺寸为224×224、格式为*.jpg的图像。数据集中70%用于培训,15%用于验证,15%用于测试。
35
+ ### 目录
36
+ 此数据集包含三个文件夹:
37
+ - train (15000 张图像)
38
+ - test (3000 张图像)
39
+ - validation (3000 张图像)
40
+ ### 数据收集
41
+ 这个数据集中的图像是我们为一个项目从蔬菜农场和市场收集的。
42
+ ### 制作元数据文件
43
+ 运行下面`python`的代码,就可以在桌面生成三个csv格式的元数据文件、一个分类数据文件(需要放入到数据文件中)
44
+ ```python
45
+ #!/usr/bin/env python3
46
+ # -*- coding: utf-8 -*-
47
+
48
+ """
49
+ 1.下载的数据文件 Vegetable Images.zip ,并解压到桌面
50
+ 2.然后执行 python generate.py 即可生成三个元数据文件和一个分类数据文件
51
+ """
52
+ import os
53
+ from pathlib import Path
54
+
55
+
56
+ category_dict = {
57
+ 'Bean': '豆类',
58
+ 'Bitter_Gourd': '苦瓜',
59
+ 'Bottle_Gourd': '葫芦',
60
+ 'Brinjal': '茄子',
61
+ 'Broccoli': '西兰花',
62
+ 'Cabbage': '卷心菜',
63
+ 'Capsicum': '辣椒',
64
+ 'Carrot': '胡萝卜',
65
+ 'Cauliflower': '花椰菜',
66
+ 'Cucumber': '黄瓜',
67
+ 'Papaya': '木瓜',
68
+ 'Potato': '土豆',
69
+ 'Pumpkin': '南瓜',
70
+ 'Radish': '萝卜',
71
+ 'Tomato': '番茄',
72
+ }
73
+ base_path = Path.home().joinpath('desktop')
74
+ data = '\n'.join((item for item in category_dict.values())) # 注意:利用了python 3.6之后字典插入有序的特性
75
+ base_path.joinpath('classname.txt').write_text(data, encoding='utf-8')
76
+
77
+
78
+ def create(filename):
79
+ csv_path = base_path.joinpath(f'{filename}.csv')
80
+ with csv_path.open('wt', encoding='utf-8', newline='') as csv:
81
+ csv.writelines([f'image,category{os.linesep}'])
82
+ data_path = base_path.joinpath('Vegetable Images', filename)
83
+ batch = 0
84
+ datas = []
85
+ keys = list(category_dict.keys())
86
+ for image_path in data_path.rglob('*.jpg'):
87
+ batch += 1
88
+ part1 = str(image_path).removeprefix(str(base_path)).replace('\\', '/')[1:]
89
+ part2 = keys.index(image_path.parents[0].name)
90
+ datas.append(f'{part1},{part2}{os.linesep}')
91
+ if batch > 100:
92
+ csv.writelines(datas)
93
+ datas.clear()
94
+ if datas:
95
+ csv.writelines(datas)
96
+ return csv_path.stat().st_size
97
+
98
+
99
+ if __name__ == '__main__':
100
+ print(create('train'))
101
+ print(create('test'))
102
+ print(create('validation'))
103
+ ```
104
+ ### 致谢
105
+ 非常感谢原始数据集提供方 [Vegetable Image Dataset](https://www.kaggle.com/datasets/misrakahmed/vegetable-image-dataset)。
106
+ ### 克隆数据
107
+ ```bash
108
+ git clone https://huggingface.co/datasets/cc92yy3344/vegetable.git
109
+ ```