图像标注格式转换器(image-annotations)
使用方法
from image_annotations import yolo2coco, yolo2voc # YOLO格式转COCO格式, YOLO格式转VOC格式 from image_annotations import voc2coco, voc2yolo # VOC格式转COCO格式, VOC格式转YOLO格式 from image_annotations import coco2yolo, coco2voc # COCO格式转YOLO格式, COCO格式转VOC格式 from image_annotations import yolo_classes_ids, coco_classes, voc_classes # 获取YOLO、COCO、VOC格式的所有类别根据需要导入相关函数,按照函数签名及注释填入相关参数即可。参数中
**_path表示文件的路径,**_dir表示文件夹路径,均推荐使用绝对路径。
标注文件说明
| 标注格式 | 文件类型 | 一张图片对应 | 一个检测框对应 |
|---|---|---|---|
| YOLO | txt | 一个txt文件 | txt文件中的一行 |
| VOC | xml | 一个xml文件 | xml文件中的一个object标签 |
| COCO | json | json文件中images的一项 | json文件中annotations的一项 |
YOLO格式
YOLO格式的目标检测标注文件通常以txt文件给出,一个txt标注文件对应一张图片,txt文件的一行对应一个目标检测框。标注格式如下:class_id x_center/width y_center/height w/width h/height其中,class_id表示类别ID,x_center表示标注框中心点x坐标,y_center表示标注框中心点y坐标,w表示标注框宽度,h表示标注框高度,width表示图片宽度,height表示图片高度。
txt标注文件示例如下:
2 0.079166666 0.6759259 0.090625 0.11666667 1 0.22552083 0.67314816 0.015625 0.048148148 1 0.21484375 0.6759259 0.0140625 0.04074074 1 0.1890625 0.6726852 0.016666668 0.047222223 1 0.17916666 0.67083335 0.014583333 0.049074072 1 0.15520833 0.6712963 0.015625 0.05
VOC格式
VOC格式的目标检测文件通常以xml文件给出,一个xml标注文件对应一张图片,xml文件的一个object标签对应一个目标检测框。xml标注文件示例如下:
<annotation> <folder>Desktop</folder> <filename>test.jpg</filename> <path>/home/DrZon/test.jpg</path> <source> <database>Unknown</database> </source> <size> <width>194</width> <height>259</height> <depth>3</depth> </size> <segmented>0</segmented> <object> <name>categoryName</name> <pose>Unspecified</pose> <truncated>0</truncated> <difficult>0</difficult> <bndbox> <xmin>56</xmin> <ymin>22</ymin> <xmax>132</xmax> <ymax>229</ymax> </bndbox> </object> </annotation>
COCO格式
COCO格式的目标检测标注文件通常以json文件给出,将所有图片的所有标注写在同一个文件里面,示例格式如下:{
"info": {
"year": 2024,
"version": "1.0",
"description": "目标检测训练数据集",
"contributor": "Your Name",
"url": "",
"date_created": "2024-06-15"
},
"licenses": [{
"id": 1,
"name": "Academic Use Only",
"url": ""
}],
"images": [
{
"id": 1,
"license": 1,
"file_name": "000001.jpg",
"height": 600,
"width": 800,
"date_captured": "2024-06-15 10:30:00"
},
{
"id": 2,
"license": 1,
"file_name": "000002.jpg",
"height": 480,
"width": 640,
"date_captured": "2024-06-15 10:31:00"
}
],
"annotations": [
{
"id": 1,
"image_id": 1,
"category_id": 1,
"bbox": [120, 150, 80, 120],
"area": 9600,
"segmentation": [],
"iscrowd": 0
},
{
"id": 2,
"image_id": 1,
"category_id": 2,
"bbox": [350, 200, 100, 60],
"area": 6000,
"segmentation": [],
"iscrowd": 0
},
{
"id": 3,
"image_id": 2,
"category_id": 1,
"bbox": [50, 80, 60, 100],
"area": 6000,
"segmentation": [],
"iscrowd": 0
}
],
"categories": [
{
"id": 1,
"name": "person",
"supercategory": "human"
},
{
"id": 2,
"name": "car",
"supercategory": "vehicle"
}
]
}
更新日志
| 版本 | 更新内容 | 更新日期 |
|---|---|---|
| 0.1.0 | 实现YOLO、COCO、VOC三种格式的目标检测标注文件相互转换 | 2026年1月6日 |
| 0.2.0 | 新增获取全部类名的功能 | 2026年1月7日 |
| 0.2.1 | Updated YOLO to VOC and COCO conversion functions to iterate over image files instead of annotation files, improving robustness for various image formats. | 2026年1月11日 |
| 0.2.2 | Exposes conversion functions between YOLO, VOC, and COCO formats, as well as class retrieval utilities, at the package level. | 2026年1月12日 |
Metadata
Release files for image-annotations 0.2.2
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| image_annotations-0.2.2.tar.gz | 23.2 kB | Without provenance | Details for image_annotations-0.2.2.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| image_annotations-0.2.2-py3-none-any.whl | Python 3 | none | any | Details for image_annotations-0.2.2-py3-none-any.whl |
Total release size: 47.3 kB
Release files / image_annotations-0.2.2.tar.gz
| Download URL | image_annotations-0.2.2.tar.gz |
|---|---|
| Size | 23.2 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
e8b6676319055b7a361f2022741504d856fd1609ed301ad1bec362fec74d5753
|
|
BLAKE2b-256 checksum How to use checksums |
b914bd420b88698a1879215d60f138f50e289f9200495745680a608d66cea234
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.10.11
|
Release files / image_annotations-0.2.2-py3-none-any.whl
| Download URL | image_annotations-0.2.2-py3-none-any.whl |
|---|---|
| Size | 24.1 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
6cb794a821d341c72811fc3f7b5f7948d217a8389bd9a7c1cd2d4212905a15b7
|
|
BLAKE2b-256 checksum How to use checksums |
66c188181e73a56a465f1d1d3303b5a4863fc13ad1bea14567c3354ea439312c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.10.11
|
Security for image-annotations 0.2.2
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.