涨分神器-基于PPYOLOE的切图和拼图解决方案
2024-06-02 19:16:54
0 项目背景
目标检测和实例分割是迄今为止计算机视觉中最重要的应用领域。然而，小物体的检测和大图像的推理仍然是实际使用中的主要问题。
比如当图像分辨率很大且样本中小目标居多时，如果reshape成小图再送进网络训练的话，目标会变得非常小，检测模型识别难度很大。
然而，直接大图训练GPU显存又顶不住，太大的原图会消耗太多的cpu时间,导致极度拖慢训练时间，而且推理速度会很慢。
如果我们可以通过离线切图的形式，把原图按一定的宽高，切成很多个小图进行训练，就能够将上面这种复杂的大图小目标检测任务，转换成较为普通的目标检测任务，从而提高小图上小目标的检测效果。接着，通过拼图与后处理，将子图预测结果再合并成大图，得到完整的大图推理结果。
百度飞桨针对上面这种大图小目标检测的典型场景，提供了PP-YOLOE Smalldet一键实现切图配置与训练。
详细文档可参考：PP-YOLOE Smalldet 检测模型。
那么，在实际业务场景中，切图拼图的目标检测任务究竟表现如何？本文就以正在进行的兴智杯国产开发框架工程化应用赛：齿轮瑕疵检测任务为例，验证自动切图训练在真实场景的表现。
1 环境准备
1.1 数据集准备
数据集的分析和准备过程在项目
兴智杯国产开发框架工程化应用赛：齿轮瑕疵检测基线中有详细的介绍，这里就不再赘述。
总体结论就是，齿轮瑕疵数据集存在密集小目标，但也有大面积的瑕疵目标。因此，正好可以被我们拿来练练手，比对下切图前后的训练效果。
# 数据集解压缩：读者可以将数据集上传到AI Studio，再根据实际项目的具体路径，解压数据集
# 注意由于数据集文件名是中文，解压的时候要指定编码（也可以本地对数据集改名后再上传）
!unzip -O GBK data/data163113/齿轮检测数据集.zip -d ./data/
# 整理数据集结构，移除脏数据
!mv data/齿轮检测数据集/train/train_coco.json data/齿轮检测数据集/
!rm data/齿轮检测数据集/train/Thumbs.db
# 引入PaddleX
!pip install paddlex
# 安装切图工具
!pip install sahi
# 组织数据目录
!mkdir MyDataset
!mkdir MyDataset/JPEGImages
!mv data/齿轮检测数据集/train/*.jpg MyDataset/JPEGImages/
!mv data/齿轮检测数据集/train_coco.json MyDataset/annotations.json
# 按比例切分数据集
!paddlex --split_dataset --format COCO --dataset_dir MyDataset --val_value 0.1 --test_value 0.0
1.2 训练环境准备
由于PP-YOLOE还在快速迭代中，因此，对框架的稳定性有一定的要求，PaddlePaddle的框架不要选择最新版。本文使用的单卡训练环境如下：
- 框架版本：PaddlePaddle 2.2.2
- CUDA Version: 11.2
- 模型库版本：PaddleDetection(release/2.5分支)
!git clone https://gitee.com/paddlepaddle/PaddleDetection.git
正克隆到 'PaddleDetection'...
remote: Enumerating objects: 27803, done.[K
remote: Counting objects: 100% (8273/8273), done.[K
remote: Compressing objects: 100% (3341/3341), done.[K
接收对象中: 17% (4876/27803), 14.54 MiB | 1.33 MiB/s
%cd PaddleDetection
/home/aistudio/PaddleDetection
# 切换到release/2.5分支
!git checkout release/2.5
M configs/smalldet/_base_/visdrone_sliced_640_025_detection.yml
M configs/smalldet/ppyoloe_crn_l_80e_sliced_visdrone_640_025.yml
M configs/smalldet/ppyoloe_crn_l_80e_sliced_visdrone_640_025_slice_infer.yml
M tools/box_distribution.py
已经位于 'release/2.5'
您的分支与上游分支 'origin/release/2.5' 一致。
1.3 切图数据准备
1.3.1 统计数据集分布
在切图前，我们首先需要统计所用数据集标注框的平均宽高占图片真实宽高的比例分布：
--json_path：待统计数据集COCO 格式 annotation 的json文件路径--out_img：输出的统计分布图路径
# PaddleDetection中box_distribution.py文件有点小bug，这里先进行修复
!cp ../box_distribution.py tools/box_distribution.py
!python tools/box_distribution.py --json_path /home/aistudio/MyDataset/annotations.json --out_img box_distribution.jpg
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/numpy/core/fromnumeric.py:3420: RuntimeWarning: Mean of empty slice.out=out, **kwargs)
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/numpy/core/_methods.py:188: RuntimeWarning: invalid value encountered in double_scalarsret = ret.dtype.type(ret / rcount)
Median of ratio_w is 0.050386904761904765
Median of ratio_h is 0.0305
all_img with box: 1399
all_ann: 28575
Distribution saved as box_distribution.jpg
Figure(640x480)
注意:
- 当原始数据集全部有标注框的图片中，有1/2以上的图片标注框的平均宽高与原图宽高比例小于0.04时，建议进行切图训练。
所以根据上面的说明，齿轮瑕疵检测这个数据集其实介于“可切可不切”之间。当然，我们可以继续试验下，看看切图后能达到的效果。
1.3.2 基于SAHI切图
针对需要切图的数据集，使用SAHI库进行切分：
--image_dir：原始数据集图片文件夹的路径--json_path：原始数据集COCO格式的json标注文件的路径--output_dir：切分后的子图及其json标注文件保存的路径--slice_size：切分以后子图的边长尺度大小(默认切图后为正方形)--overlap_ratio：切分时的子图之间的重叠率- 以上述代码为例，切分后的子图文件夹与json标注文件共同保存在
MyDataset/IMG_sliced文件夹下，比如训练集图片和标注就命名为train_images_640_025、train_images_640_025.json。
!mkdir /home/aistudio/MyDataset/IMG_sliced
# 对训练集标注进行切图
!python tools/slice_image.py --image_dir /home/aistudio/MyDataset/JPEGImages --json_path /home/aistudio/MyDataset/train.json --output_dir /home/aistudio/MyDataset/IMG_sliced --slice_size 640 --overlap_ratio 0.25
# 对验证集标注进行切图
!python tools/slice_image.py --image_dir /home/aistudio/MyDataset/JPEGImages --json_path /home/aistudio/MyDataset/val.json --output_dir /home/aistudio/MyDataset/IMG_sliced --slice_size 640 --overlap_ratio 0.25
2 模型训练
2.1 拼图模型选型
PaddleDetection团队提供的基于PP-YOLOE的检测模型，以及提供了一套使用SAHI(Slicing Aided Hyper Inference)工具切图和拼图的方案，其效果如下：
| 模型 | 数据集 | SLICE_SIZE | OVERLAP_RATIO | 类别数 |
mAPval 0.5:0.95 |
APval 0.5 |
下载链接 | 配置文件 |
|---|---|---|---|---|---|---|---|---|
| PP-YOLOE-l | VisDrone-DET | 640 | 0.25 | 10 | 29.7 | 48.5 | 下载链接 | 配置文件 |
| PP-YOLOE-l (Assembled) | VisDrone-DET | 640 | 0.25 | 10 | 37.2 | 59.4 | 下载链接 | 配置文件 |
Assembled表示自动切图和拼图后模型的表现，从中我们可以看出，mAP较原图预测有了非常显著的提升，因此，接下来就基于PP-YOLOE-l，看看自动切图和拼图后模型在齿轮瑕疵检测数据集上的表现。
2.2 模型训练
毫无疑问，切图后模型的训练是要基于切图数据集的，配置如下：
TrainDataset:!COCODataSetimage_dir: /home/aistudio/MyDataset/IMG_sliced/train_images_640_025anno_path: /home/aistudio/MyDataset/IMG_sliced/train_640_025.jsondataset_dir: /home/aistudio/MyDatasetdata_fields: ['image', 'gt_bbox', 'gt_class', 'is_crowd']EvalDataset:!COCODataSetimage_dir: /home/aistudio/MyDataset/IMG_sliced/val_images_640_025anno_path: /home/aistudio/MyDataset/IMG_sliced/val_640_025.jsondataset_dir: /home/aistudio/MyDataset
# 覆盖配置文件
!cp ../ppyoloe_crn_l_80e_sliced_visdrone_640_025.yml configs/smalldet/ppyoloe_crn_l_80e_sliced_visdrone_640_025.yml
!cp ../visdrone_sliced_640_025_detection.yml configs/smalldet/_base_/visdrone_sliced_640_025_detection.yml
# 开始训练
!python tools/train.py -c configs/smalldet/ppyoloe_crn_l_80e_sliced_visdrone_640_025.yml --use_vdl=True --vdl_log_dir=./sliced_visdrone/ --eval
切图训练时间是否一定比原图训练时间长？这个要视情况而定。
因为切图后训练数据集的数据量虽然大幅增加了，但是切图后输入图片可以减小分辨率（从而增大batchsize）。当然通常来说，切图训练时间要比原图训练时间要长一些。
在原图训练的PPYOLOE：又快又好的小目标检测训练与部署实现中，10个epoch训练耗时在60min左右；切图训练则需要48min左右，因为增大了batchsize，训练效率反而提高了。
而且，切图模型收敛速度相较于原图训练，明显快了很多。
3 模型评估
我们对训练30个epoch后，切图模型的效果进行一下评估。
3.1 子图评估
对于子图评估和原图评估，差别仅仅在于验证集路径的配置：
- 子图评估
- 配置切图后的子图存放目录和子图验证集标注文件
- 原图评估
- 配置原图存放目录和验证集标注文件
# 训练30个epoch后，子图评估
!python tools/eval.py -c configs/smalldet/ppyoloe_crn_l_80e_sliced_visdrone_640_025.yml -o weights=output/ppyoloe_crn_l_80e_sliced_visdrone_640_025/best_model.pdparams
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:36: DeprecationWarning: NEAREST is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.NEAREST or Dither.NONE instead.'nearest': Image.NEAREST,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:37: DeprecationWarning: BILINEAR is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BILINEAR instead.'bilinear': Image.BILINEAR,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:38: DeprecationWarning: BICUBIC is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BICUBIC instead.'bicubic': Image.BICUBIC,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:39: DeprecationWarning: BOX is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BOX instead.'box': Image.BOX,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:40: DeprecationWarning: LANCZOS is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.LANCZOS instead.'lanczos': Image.LANCZOS,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:41: DeprecationWarning: HAMMING is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.HAMMING instead.'hamming': Image.HAMMING
Warning: Unable to use OC-SORT, please install filterpy, for example: `pip install filterpy`, see https://github.com/rlabbe/filterpy
Warning: import ppdet from source directory without installing, run 'python setup.py install' to install ppdet firstly
W0906 00:49:25.085027 31458 device_context.cc:447] Please NOTE: device: 0, GPU Compute Capability: 7.0, Driver API Version: 11.2, Runtime API Version: 10.1
W0906 00:49:25.089991 31458 device_context.cc:465] device: 0, cuDNN Version: 7.6.
loading annotations into memory...
Done (t=0.02s)
creating index...
index created!
[09/06 00:49:30] ppdet.utils.checkpoint INFO: Finish loading model weights: output/ppyoloe_crn_l_80e_sliced_visdrone_640_025/best_model.pdparams
[09/06 00:49:30] ppdet.engine INFO: Eval iter: 0
[09/06 00:49:37] ppdet.engine INFO: Eval iter: 100
[09/06 00:49:44] ppdet.engine INFO: Eval iter: 200
[09/06 00:49:51] ppdet.engine INFO: Eval iter: 300
[09/06 00:49:58] ppdet.engine INFO: Eval iter: 400
[09/06 00:50:05] ppdet.metrics.metrics INFO: The bbox result is saved to bbox.json.
loading annotations into memory...
Done (t=0.03s)
creating index...
index created!
[09/06 00:50:05] ppdet.metrics.coco_utils INFO: Start evaluate...
Loading and preparing results...
DONE (t=3.95s)
creating index...
index created!
Running per image evaluation...
Evaluate annotation type *bbox*
DONE (t=13.39s).
Accumulating evaluation results...
DONE (t=1.71s).Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.490Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.816Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.522Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.459Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.543Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.611Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.130Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.481Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.639Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.535Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.688Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.874
[09/06 00:50:25] ppdet.engine INFO: Total sample number: 812, averge FPS: 27.543498235955227
3.2 原图评估
!cp ../ppyoloe_crn_l_80e_sliced_visdrone_640_025-Copy1.yml configs/smalldet/ppyoloe_crn_l_80e_sliced_visdrone_640_025-Copy1.yml
# 训练30个epoch后，原图评估
!python tools/eval.py -c configs/smalldet/ppyoloe_crn_l_80e_sliced_visdrone_640_025-Copy1.yml -o weights=output/ppyoloe_crn_l_80e_sliced_visdrone_640_025/best_model.pdparams
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:36: DeprecationWarning: NEAREST is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.NEAREST or Dither.NONE instead.'nearest': Image.NEAREST,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:37: DeprecationWarning: BILINEAR is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BILINEAR instead.'bilinear': Image.BILINEAR,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:38: DeprecationWarning: BICUBIC is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BICUBIC instead.'bicubic': Image.BICUBIC,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:39: DeprecationWarning: BOX is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BOX instead.'box': Image.BOX,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:40: DeprecationWarning: LANCZOS is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.LANCZOS instead.'lanczos': Image.LANCZOS,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:41: DeprecationWarning: HAMMING is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.HAMMING instead.'hamming': Image.HAMMING
Warning: Unable to use OC-SORT, please install filterpy, for example: `pip install filterpy`, see https://github.com/rlabbe/filterpy
Warning: import ppdet from source directory without installing, run 'python setup.py install' to install ppdet firstly
W0906 00:54:40.913517 32345 device_context.cc:447] Please NOTE: device: 0, GPU Compute Capability: 7.0, Driver API Version: 11.2, Runtime API Version: 10.1
W0906 00:54:40.918522 32345 device_context.cc:465] device: 0, cuDNN Version: 7.6.
loading annotations into memory...
Done (t=0.02s)
creating index...
index created!
[09/06 00:54:45] ppdet.utils.checkpoint INFO: Finish loading model weights: output/ppyoloe_crn_l_80e_sliced_visdrone_640_025/best_model.pdparams
[09/06 00:54:46] ppdet.engine INFO: Eval iter: 0
[09/06 00:54:55] ppdet.metrics.metrics INFO: The bbox result is saved to bbox.json.
loading annotations into memory...
Done (t=0.01s)
creating index...
index created!
[09/06 00:54:55] ppdet.metrics.coco_utils INFO: Start evaluate...
Loading and preparing results...
DONE (t=0.65s)
creating index...
index created!
Running per image evaluation...
Evaluate annotation type *bbox*
DONE (t=5.27s).
Accumulating evaluation results...
DONE (t=0.23s).Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.389Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.733Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.361Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.213Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.384Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.567Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.050Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.276Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.520Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.299Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.505Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.679
[09/06 00:55:01] ppdet.engine INFO: Total sample number: 141, averge FPS: 17.53671755623329
3.3 子图拼图评估
对于子图拼图评估，需要修改验证集的标注文件路径为原图标注文件，然后配置设置如下：
EvalDataset:!SlicedCOCODataSetimage_dir: VisDrone2019-DET-valanno_path: val.jsondataset_dir: dataset/visdronesliced_size: [640, 640]overlap_ratio: [0.25, 0.25]
需要注意的是，子图拼图评估耗时可能非常长，它取决于模型权重文件的大小。同时，在配置子图拼图评估是，需要用SlicedCOCODataSet替代COCODataSet，并且配置切图训练时的sliced_size和overlap_ratio参数：
- 设置
--slice_infer表示切图预测并拼装重组结果，如果不使用则不写，注意需要确保EvalDataset的数据集类是选用的SlicedCOCODataSet而不是COCODataSet； - 设置
--slice_size表示切图的子图尺寸大小，设置--overlap_ratio表示子图间重叠率，可以自行修改选择合适的子图尺度sliced_size和子图间重叠率overlap_ratio。
!cp ../ppyoloe_crn_l_80e_sliced_visdrone_640_025_slice_infer.yml configs/smalldet/ppyoloe_crn_l_80e_sliced_visdrone_640_025_slice_infer.yml
执行命令时，需要关注下面这些参数的设置：
- 设置
--combine_method表示子图结果重组去重的方式，默认是nms； - 设置
--match_threshold表示子图结果重组去重的阈值，默认是0.6； - 设置
--match_metric表示子图结果重组去重的度量标准，默认是ios表示交小比(两个框交集面积除以更小框的面积)，也可以选择交并比iou(两个框交集面积除以并集面积)，精度效果因数据集而而异，但选择ios预测速度会更快一点。
在本项目中，我们把交小比和交并比两种方式都试验一番：
# 训练30个epoch后，子图拼图评估，交小比
!python tools/eval.py -c configs/smalldet/ppyoloe_crn_l_80e_sliced_visdrone_640_025_slice_infer.yml -o weights=output/ppyoloe_crn_l_80e_sliced_visdrone_640_025/best_model.pdparams --slice_infer --combine_method=nms --match_threshold=0.6 --match_metric=ios
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:36: DeprecationWarning: NEAREST is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.NEAREST or Dither.NONE instead.'nearest': Image.NEAREST,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:37: DeprecationWarning: BILINEAR is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BILINEAR instead.'bilinear': Image.BILINEAR,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:38: DeprecationWarning: BICUBIC is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BICUBIC instead.'bicubic': Image.BICUBIC,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:39: DeprecationWarning: BOX is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BOX instead.'box': Image.BOX,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:40: DeprecationWarning: LANCZOS is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.LANCZOS instead.'lanczos': Image.LANCZOS,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:41: DeprecationWarning: HAMMING is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.HAMMING instead.'hamming': Image.HAMMING
Warning: Unable to use OC-SORT, please install filterpy, for example: `pip install filterpy`, see https://github.com/rlabbe/filterpy
Warning: import ppdet from source directory without installing, run 'python setup.py install' to install ppdet firstly
W0906 00:58:13.224578 584 device_context.cc:447] Please NOTE: device: 0, GPU Compute Capability: 7.0, Driver API Version: 11.2, Runtime API Version: 10.1
W0906 00:58:13.229662 584 device_context.cc:465] device: 0, cuDNN Version: 7.6.
loading annotations into memory...
Done (t=0.02s)
creating index...
index created!
[09/06 00:58:23] ppdet.data.source.coco INFO: 1932 samples and slice to 1932 sub_samples in file /home/aistudio/MyDataset/IMG_sliced/val_640_025.json
[09/06 00:58:24] ppdet.utils.checkpoint INFO: Finish loading model weights: output/ppyoloe_crn_l_80e_sliced_visdrone_640_025/best_model.pdparams
[09/06 00:58:25] ppdet.engine INFO: Eval iter: 0
[09/06 00:58:35] ppdet.engine INFO: Eval iter: 100
[09/06 00:58:45] ppdet.engine INFO: Eval iter: 200
[09/06 00:58:57] ppdet.engine INFO: Eval iter: 300
[09/06 00:59:07] ppdet.engine INFO: Eval iter: 400
[09/06 00:59:19] ppdet.engine INFO: Eval iter: 500
[09/06 00:59:31] ppdet.engine INFO: Eval iter: 600
[09/06 00:59:42] ppdet.engine INFO: Eval iter: 700
[09/06 00:59:52] ppdet.engine INFO: Eval iter: 800
[09/06 01:00:02] ppdet.engine INFO: Eval iter: 900
[09/06 01:00:13] ppdet.engine INFO: Eval iter: 1000
[09/06 01:00:24] ppdet.engine INFO: Eval iter: 1100
[09/06 01:00:36] ppdet.engine INFO: Eval iter: 1200
[09/06 01:00:46] ppdet.engine INFO: Eval iter: 1300
[09/06 01:00:58] ppdet.engine INFO: Eval iter: 1400
[09/06 01:01:09] ppdet.engine INFO: Eval iter: 1500
[09/06 01:01:19] ppdet.engine INFO: Eval iter: 1600
[09/06 01:01:30] ppdet.engine INFO: Eval iter: 1700
[09/06 01:01:41] ppdet.engine INFO: Eval iter: 1800
[09/06 01:01:52] ppdet.engine INFO: Eval iter: 1900
[09/06 01:01:59] ppdet.metrics.metrics INFO: The bbox result is saved to bbox.json.
loading annotations into memory...
Done (t=0.02s)
creating index...
index created!
[09/06 01:01:59] ppdet.metrics.coco_utils INFO: Start evaluate...
Loading and preparing results...
DONE (t=2.22s)
creating index...
index created!
Running per image evaluation...
Evaluate annotation type *bbox*
DONE (t=9.34s).
Accumulating evaluation results...
DONE (t=2.21s).Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.419Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.711Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.444Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.399Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.458Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.474Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.130Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.442Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.517Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.452Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.535Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.701
[09/06 01:02:13] ppdet.engine INFO: Total sample number: 1932, averge FPS: 9.132096605227849
# 训练30个epoch后，子图拼图评估，交并比
!python tools/eval.py -c configs/smalldet/ppyoloe_crn_l_80e_sliced_visdrone_640_025_slice_infer.yml -o weights=output/ppyoloe_crn_l_80e_sliced_visdrone_640_025/best_model.pdparams --slice_infer --combine_method=nms --match_threshold=0.6 --match_metric=iou
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:36: DeprecationWarning: NEAREST is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.NEAREST or Dither.NONE instead.'nearest': Image.NEAREST,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:37: DeprecationWarning: BILINEAR is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BILINEAR instead.'bilinear': Image.BILINEAR,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:38: DeprecationWarning: BICUBIC is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BICUBIC instead.'bicubic': Image.BICUBIC,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:39: DeprecationWarning: BOX is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BOX instead.'box': Image.BOX,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:40: DeprecationWarning: LANCZOS is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.LANCZOS instead.'lanczos': Image.LANCZOS,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:41: DeprecationWarning: HAMMING is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.HAMMING instead.'hamming': Image.HAMMING
Warning: Unable to use OC-SORT, please install filterpy, for example: `pip install filterpy`, see https://github.com/rlabbe/filterpy
Warning: import ppdet from source directory without installing, run 'python setup.py install' to install ppdet firstly
W0906 01:02:17.801489 1241 device_context.cc:447] Please NOTE: device: 0, GPU Compute Capability: 7.0, Driver API Version: 11.2, Runtime API Version: 10.1
W0906 01:02:17.806488 1241 device_context.cc:465] device: 0, cuDNN Version: 7.6.
loading annotations into memory...
Done (t=0.02s)
creating index...
index created!
[09/06 01:02:27] ppdet.data.source.coco INFO: 1932 samples and slice to 1932 sub_samples in file /home/aistudio/MyDataset/IMG_sliced/val_640_025.json
[09/06 01:02:29] ppdet.utils.checkpoint INFO: Finish loading model weights: output/ppyoloe_crn_l_80e_sliced_visdrone_640_025/best_model.pdparams
[09/06 01:02:30] ppdet.engine INFO: Eval iter: 0
[09/06 01:03:10] ppdet.engine INFO: Eval iter: 100
[09/06 01:04:02] ppdet.engine INFO: Eval iter: 200
[09/06 01:04:47] ppdet.engine INFO: Eval iter: 300
[09/06 01:05:27] ppdet.engine INFO: Eval iter: 400
[09/06 01:06:14] ppdet.engine INFO: Eval iter: 500
[09/06 01:07:05] ppdet.engine INFO: Eval iter: 600
[09/06 01:07:55] ppdet.engine INFO: Eval iter: 700
[09/06 01:08:28] ppdet.engine INFO: Eval iter: 800
[09/06 01:09:18] ppdet.engine INFO: Eval iter: 900
[09/06 01:10:03] ppdet.engine INFO: Eval iter: 1000
[09/06 01:10:50] ppdet.engine INFO: Eval iter: 1100
[09/06 01:11:35] ppdet.engine INFO: Eval iter: 1200
[09/06 01:12:14] ppdet.engine INFO: Eval iter: 1300
[09/06 01:13:01] ppdet.engine INFO: Eval iter: 1400
[09/06 01:13:45] ppdet.engine INFO: Eval iter: 1500
[09/06 01:14:24] ppdet.engine INFO: Eval iter: 1600
[09/06 01:15:08] ppdet.engine INFO: Eval iter: 1700
[09/06 01:15:49] ppdet.engine INFO: Eval iter: 1800
[09/06 01:16:31] ppdet.engine INFO: Eval iter: 1900
[09/06 01:16:57] ppdet.metrics.metrics INFO: The bbox result is saved to bbox.json.
loading annotations into memory...
Done (t=0.02s)
creating index...
index created!
[09/06 01:16:58] ppdet.metrics.coco_utils INFO: Start evaluate...
Loading and preparing results...
DONE (t=7.08s)
creating index...
index created!
Running per image evaluation...
Evaluate annotation type *bbox*
DONE (t=16.24s).
Accumulating evaluation results...
DONE (t=4.36s).Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.471Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.785Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.502Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.443Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.532Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.567Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.130Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.480Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.635Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.530Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.684Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.870
[09/06 01:17:27] ppdet.engine INFO: Total sample number: 1932, averge FPS: 2.256297048751735
3.4 评估结论
将30个epoch后，原图训练、切图训练子图评估、切图训练原图评估、切图训练子图拼图评估的效果进行了对比，结果如下：
| 训练评估方式 | SLICE_SIZE | OVERLAP_RATIO |
mAPval 0.5:0.95 |
APval 0.5 |
|---|---|---|---|---|
| 原图训练 | - | - | 33.8 | 64.0 |
| 子图训练子图评估 | 640 | 0.25 | 49.0 | 81.6 |
| 子图训练原图评估 | 640 | 0.25 | 38.9 | 73.3 |
| 子图训练拼图评估 - IoS | 640 | 0.25 | 41.9 | 78.5 |
| 子图训练拼图评估 - IoU | 640 | 0.25 | 47.1 | 71.1 |
从上面的简单表格可以明显看出，相比原图直接训练，子图训练拼图评估精度提升明显。同时，子图结果重组去重的度量标准用交并比在该数据集上表现更好，猜测可能是因为，数据集的目标大小差异较大，对于中大型目标，去重标准如果用交小比，在评估效果上，会比较吃亏。
4 拼图预测
# 准备一个放测试集图片的目录，然后将待预测的示例图片移至该目录下
!unzip -O GBK ../data/data163113/齿轮检测A榜评测数据.zip -d ../data/
!mkdir ../data/test
!mv ../data/齿轮检测A榜评测数据/val/*.jpg ../data/test/
4.1 单图切图拼图预测
# 挑一张测试集的图片展示预测效果
!python tools/infer.py -c configs/smalldet/ppyoloe_crn_l_80e_sliced_visdrone_640_025.yml -o weights=output/ppyoloe_crn_l_80e_sliced_visdrone_640_025/best_model.pdparams --infer_img=../data/test/1__H2_817171_IO-NIO198M_210121A0050-1-1.jpg --draw_threshold=0.25 --slice_infer --slice_size 640 640 --overlap_ratio 0.25 0.25 --combine_method=nms --match_threshold=0.6 --match_metric=iou --save_results=True
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:36: DeprecationWarning: NEAREST is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.NEAREST or Dither.NONE instead.'nearest': Image.NEAREST,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:37: DeprecationWarning: BILINEAR is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BILINEAR instead.'bilinear': Image.BILINEAR,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:38: DeprecationWarning: BICUBIC is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BICUBIC instead.'bicubic': Image.BICUBIC,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:39: DeprecationWarning: BOX is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BOX instead.'box': Image.BOX,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:40: DeprecationWarning: LANCZOS is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.LANCZOS instead.'lanczos': Image.LANCZOS,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:41: DeprecationWarning: HAMMING is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.HAMMING instead.'hamming': Image.HAMMING
Warning: Unable to use OC-SORT, please install filterpy, for example: `pip install filterpy`, see https://github.com/rlabbe/filterpy
Warning: import ppdet from source directory without installing, run 'python setup.py install' to install ppdet firstly
W0906 14:35:08.361063 26272 device_context.cc:447] Please NOTE: device: 0, GPU Compute Capability: 7.0, Driver API Version: 11.2, Runtime API Version: 10.1
W0906 14:35:08.364892 26272 device_context.cc:465] device: 0, cuDNN Version: 7.6.
[09/06 14:35:12] ppdet.utils.checkpoint INFO: Finish loading model weights: output/ppyoloe_crn_l_80e_sliced_visdrone_640_025/best_model.pdparams
1 samples and slice to 12 sub_samples
loading annotations into memory...
Done (t=0.01s)
creating index...
index created!
100%|███████████████████████████████████████████| 12/12 [00:35<00:00, 2.92s/it]
[09/06 14:35:47] ppdet.engine INFO: Detection bbox results save in output/1__H2_817171_IO-NIO198M_210121A0050-1-1.jpg
感兴趣的读者可以把这张测试图片的预测效果与PPYOLOE：又快又好的小目标检测训练与部署实现中推理预测的同一张图片进行对比，肉眼可见的是：“抓到的目标”变多了！
当然，一张图片的预测就执行了44s，这属于典型的以时间换精度了！
4.2 批量拼图预测
# 执行批量预测，当前还不支持生成含有预测结果的bbox.json文件
!python tools/infer.py -c configs/smalldet/ppyoloe_crn_l_80e_sliced_visdrone_640_025.yml -o weights=output/ppyoloe_crn_l_80e_sliced_visdrone_640_025/best_model.pdparams --infer_dir=../data/test --slice_infer --slice_size 640 640 --overlap_ratio 0.25 0.25 --combine_method=nms --match_threshold=0.6 --match_metric=ios --save_results=True
5 模型部署
5.1 导出模型
# export model
!python tools/export_model.py -c configs/smalldet/ppyoloe_crn_l_80e_sliced_visdrone_640_025.yml -o weights=output/ppyoloe_crn_l_80e_sliced_visdrone_640_025/best_model.pdparams
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:36: DeprecationWarning: NEAREST is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.NEAREST or Dither.NONE instead.'nearest': Image.NEAREST,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:37: DeprecationWarning: BILINEAR is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BILINEAR instead.'bilinear': Image.BILINEAR,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:38: DeprecationWarning: BICUBIC is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BICUBIC instead.'bicubic': Image.BICUBIC,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:39: DeprecationWarning: BOX is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BOX instead.'box': Image.BOX,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:40: DeprecationWarning: LANCZOS is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.LANCZOS instead.'lanczos': Image.LANCZOS,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:41: DeprecationWarning: HAMMING is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.HAMMING instead.'hamming': Image.HAMMING
Warning: Unable to use OC-SORT, please install filterpy, for example: `pip install filterpy`, see https://github.com/rlabbe/filterpy
Warning: import ppdet from source directory without installing, run 'python setup.py install' to install ppdet firstly
[09/06 20:38:09] ppdet.utils.checkpoint INFO: Finish loading model weights: output/ppyoloe_crn_l_80e_sliced_visdrone_640_025/best_model.pdparams
loading annotations into memory...
Done (t=0.01s)
creating index...
index created!
[09/06 20:38:10] ppdet.engine INFO: Export inference config file to output_inference/ppyoloe_crn_l_80e_sliced_visdrone_640_025/infer_cfg.yml
[09/06 20:38:18] ppdet.engine INFO: Export model and saved in output_inference/ppyoloe_crn_l_80e_sliced_visdrone_640_025
5.2 自动切图拼图推理
!python deploy/python/infer.py --model_dir=output_inference/ppyoloe_crn_l_80e_sliced_visdrone_640_025 --image_file=../data/test/1__H2_817171_IO-NIO198M_210121A0050-1-1.jpg --device=GPU --save_images=True --threshold=0.25 --slice_infer --slice_size 640 640 --overlap_ratio 0.25 0.25 --combine_method=nms --match_threshold=0.6 --match_metric=iou
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:36: DeprecationWarning: NEAREST is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.NEAREST or Dither.NONE instead.'nearest': Image.NEAREST,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:37: DeprecationWarning: BILINEAR is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BILINEAR instead.'bilinear': Image.BILINEAR,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:38: DeprecationWarning: BICUBIC is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BICUBIC instead.'bicubic': Image.BICUBIC,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:39: DeprecationWarning: BOX is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BOX instead.'box': Image.BOX,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:40: DeprecationWarning: LANCZOS is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.LANCZOS instead.'lanczos': Image.LANCZOS,
/opt/conda/envs/python35-paddle120-env/lib/python3.7/site-packages/paddle/vision/transforms/functional_pil.py:41: DeprecationWarning: HAMMING is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.HAMMING instead.'hamming': Image.HAMMING
----------- Running Arguments -----------
action_file: None
batch_size: 1
camera_id: -1
combine_method: nms
cpu_threads: 1
device: GPU
enable_mkldnn: False
enable_mkldnn_bfloat16: False
image_dir: None
image_file: ../data/test/1__H2_817171_IO-NIO198M_210121A0050-1-1.jpg
match_metric: iou
match_threshold: 0.6
model_dir: output_inference/ppyoloe_crn_l_80e_sliced_visdrone_640_025
output_dir: output
overlap_ratio: [0.25, 0.25]
random_pad: False
reid_batch_size: 50
reid_model_dir: None
run_benchmark: False
run_mode: paddle
save_images: True
save_mot_txt_per_img: False
save_mot_txts: False
save_results: False
scaled: False
slice_infer: True
slice_size: [640, 640]
threshold: 0.25
tracker_config: None
trt_calib_mode: False
trt_max_shape: 1280
trt_min_shape: 1
trt_opt_shape: 640
use_coco_category: False
use_dark: True
use_gpu: False
video_file: None
window_size: 50
------------------------------------------
----------- Model Configuration -----------
Model Arch: YOLO
Transform Order:
--transform op: Resize
--transform op: NormalizeImage
--transform op: Permute
--------------------------------------------
/home/aistudio/PaddleDetection/deploy/python/utils.py:360: DeprecationWarning: `np.int` is a deprecated alias for the builtin `int`. To silence this warning, use `int` by itself. Doing this will not modify any behavior and is safe. When replacing `np.int`, you may wish to use e.g. `np.int64` or `np.int32` to specify the precision. If you wish to review your current use, check the release note link for additional information.
Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecationssuppressed = np.zeros((ndets), dtype=np.int)
class_id:2, confidence:0.8703, left_top:[425.87,63.67],right_bottom:[485.42,105.51]
/home/aistudio/PaddleDetection/deploy/python/visualize.py:162: DeprecationWarning: textsize is deprecated and will be removed in Pillow 10 (2023-07-01). Use textbbox or textlength instead.tw, th = draw.textsize(text)
class_id:2, confidence:0.8072, left_top:[522.37,116.04],right_bottom:[557.38,140.43]
/home/aistudio/PaddleDetection/deploy/python/visualize.py:162: DeprecationWarning: textsize is deprecated and will be removed in Pillow 10 (2023-07-01). Use textbbox or textlength instead.tw, th = draw.textsize(text)
class_id:2, confidence:0.7901, left_top:[316.70,10.15],right_bottom:[351.45,45.24]
/home/aistudio/PaddleDetection/deploy/python/visualize.py:162: DeprecationWarning: textsize is deprecated and will be removed in Pillow 10 (2023-07-01). Use textbbox or textlength instead.tw, th = draw.textsize(text)
class_id:2, confidence:0.7372, left_top:[353.58,20.74],right_bottom:[390.62,58.70]
/home/aistudio/PaddleDetection/deploy/python/visualize.py:162: DeprecationWarning: textsize is deprecated and will be removed in Pillow 10 (2023-07-01). Use textbbox or textlength instead.tw, th = draw.textsize(text)
class_id:2, confidence:0.5496, left_top:[589.21,151.27],right_bottom:[639.93,187.41]
/home/aistudio/PaddleDetection/deploy/python/visualize.py:162: DeprecationWarning: textsize is deprecated and will be removed in Pillow 10 (2023-07-01). Use textbbox or textlength instead.tw, th = draw.textsize(text)
class_id:2, confidence:0.4082, left_top:[302.52,-0.14],right_bottom:[334.78,10.89]
/home/aistudio/PaddleDetection/deploy/python/visualize.py:162: DeprecationWarning: textsize is deprecated and will be removed in Pillow 10 (2023-07-01). Use textbbox or textlength instead.tw, th = draw.textsize(text)
class_id:2, confidence:0.4034, left_top:[302.91,149.87],right_bottom:[333.91,181.93]
/home/aistudio/PaddleDetection/deploy/python/visualize.py:162: DeprecationWarning: textsize is deprecated and will be removed in Pillow 10 (2023-07-01). Use textbbox or textlength instead.tw, th = draw.textsize(text)
save result to: output/1__H2_817171_IO-NIO198M_210121A0050-1-1.jpg
Test iter 0
------------------ Inference Time Info ----------------------
total_time(ms): 2048.0, img_num: 1
average latency time(ms): 2048.00, QPS: 0.488281
preprocess_time(ms): 1896.70, inference_time(ms): 151.20, postprocess_time(ms): 0.10
从推理结果我们可以看出，切图的预处理过程耗时最长；同时，推理模型和训练模型在预测结果上，还存在一定的差距。
6 小结
本项目使用PP-YOLOE smalldet的自动切图拼图技术，再次升级了齿轮瑕疵检测任务的解决方案，在PPYOLOE：又快又好的小目标检测训练与部署实现的基础上进一步提高精度。模型还支持GPU上的推理部署，对于时间要求不高、精度要求极高的大图目标检测场景，具有良好的潜在应用价值。
ngth instead.
tw, th = draw.textsize(text)
save result to: output/1__H2_817171_IO-NIO198M_210121A0050-1-1.jpg
Test iter 0
------------------ Inference Time Info ----------------------
total_time(ms): 2048.0, img_num: 1
average latency time(ms): 2048.00, QPS: 0.488281
preprocess_time(ms): 1896.70, inference_time(ms): 151.20, postprocess_time(ms): 0.10
[外链图片转存中…(img-GhUDVsrt-1662984358218)]
从推理结果我们可以看出，切图的预处理过程耗时最长；同时，推理模型和训练模型在预测结果上，还存在一定的差距。
6 小结
本项目使用PP-YOLOE smalldet的自动切图拼图技术，再次升级了齿轮瑕疵检测任务的解决方案，在PPYOLOE：又快又好的小目标检测训练与部署实现的基础上进一步提高精度。模型还支持GPU上的推理部署，对于时间要求不高、精度要求极高的大图目标检测场景，具有良好的潜在应用价值。
此文章为搬运
原项目链接
涨分神器-基于PPYOLOE的切图和拼图解决方案相关推荐
- Python——基于PIL和CV2实现自动拼图解决方案
解决方案 # python3 import cv2 from PIL import Image import numpy as np import os import shutil import th ...
- 一个PS切图神器-蓝湖
作为一个前端开发,你是否有为了切图而焦头烂额的时候,拿到一张psd的时候需要将图片细细分割,这个时候估计会在心里diss一万遍ui设计师了,没事,这个时候,其实你需要的是一款神器,PS切图神器-蓝湖 ...
- 切图标记外挂神器 Assistor PS 深入解读
首先要告诉大家一个好消息,原本 $9.99 一个月 / $99.9 一年 的 Assistor PS 现已免费下载.对于学生用户全面免费,而其他用户在今年 6 月之后就不用再购买商业许可证也可以免费使 ...
- phtoshop切图的几种方式
phtoshop切图的几种方式 作为一个前端开发者,想必多多少少要掌握一定的切图技巧,通过几天的摸索,从了解psd是什么文件到安装ps并对其简单的工具进行一点的了解继而到最后怎么样切图并将设计稿变成一 ...
- ArcGIS Server 切片数学关系阐释，小区域切图频繁出错解决方法
直击: Tile|SuperTile|Bundle的关系 1.紧凑型切片中Tile|SuperTile|Bundle关系为: 1 2 3 4 5 6 7 8 9 如果:Tile = 256 × 256 ...
- ps切图后 JAVA开发_做一个会PS切图的前端开发
系列链接 切图方法分类 PhotoShop从CS版本演变到现在的CC版本,切图功能发生了比较大的变化,我们可以把PhotoShop CS版本时的切图功能称为传统切图,而从PhotoShop CC版本开 ...
- 即插即用！视频超分中的涨点神器：iSeeBetter
CNN让超分结果更真实,GAN让超分结果更丰满,所以CNN+GAN=GOOD! 添加一个鉴别器组件就能使结果增加0.32dB,即插即用,涨点神器!是否在其他的CNN架构上也可行,还需实验验证. 题目: ...
- android 从服务端获取的图片怎么适配不同分屏幕的手机,移动端的适配|切图|标注...
年前最后一个工作日,完成这篇干货- 这篇文章的目的是想从更深的原理层去找到一些设计规范的原因,比如为什么要选用@2x的750*1334做基础设计稿? 开发拿到设计基础稿是如何操作适配的? 基础稿是如何 ...
- MAC下的切图神器！SLICY使用说明
切图作为设计师出图的最后一步,向来都是令人头疼的工作,其工作内容繁琐枯燥,但又必不可少.同时,作为IOS开发者,我们往往需要为程序准备两套图片资源,以应对低清屏和Retina屏的不同需求,随之而来的是 ...
- Mac下的切图神器——Slicy使用说明
转自Akring的博客 Slicy 一.前言 切图作为美工出图的最后一步,向来都是令人头疼的工作,其工作内容繁琐枯燥,但又必不可少.同时,作为IOS开发者,我们往往需要为程序准备两 套图片资源,以应对 ...
最新文章
- 《游戏设计师修炼之道：数据驱动的游戏设计》一2.8小结
- 结构型模式之Proxy模式
- 还在魔改Transformer结构吗？微软中山大学开源超强的视觉位置编码，涨点显著
- python特性和属性的区别_什么是属性,什么是特性,有何不同?
- 关于 C++ 中输入输出的猜想
- php 502状态码,Nginx502状态码处理
- OAuth 及 移动端鉴权调研
- linux环境手动编译安装Nginx实践过程 附异常解决
- Solaris11修改主机名
- 【坐在马桶上看算法】算法3：最常用的排序——快速排序
- 【语音处理】基于matlab GUI录音信号时域频域分析（带面板）【含Matlab源码 064期】
- 看看哪些小程序能帮你赚点零花钱！
- DICOM图像像素值、灰度值与CT值
- 高等数学张宇18讲 第六讲 零点问题、微分不等式
- 2021图像检索综述
- Eterm连接不上-10001:登录失败
- 一个非常好用的免费图床
- 5336: [TJOI2018]party
- 安装Charles证书失败（夜神模拟器总是弹出此网站安全证书有问题），对安卓手机网络进行抓包
- 莺尾花数据集–贝叶斯分类(day5)